CN105787605A - Micro-grid economic and optimal operation and scheduling method based on improved quantum genetic algorithm - Google Patents

Abstract

The invention relates to a micro-grid economic and optimal operation and scheduling method based on an improved quantum genetic algorithm. The micro-grid is in a grid-connected mode operation state and comprises multiple micro sources and loads, wherein the loads comprise electric loads and thermal loads; and the micro sources comprise a micro turbine, a wind turbine, a photovoltaic cell, a fuel cell, a storage battery and an electric vehicle. The method comprises the following steps: S1, state information of each load and each micro source in the micro-grid is acquired; S2, with minimum of operation cost and pollutant treatment cost as a target, a multi-target economic scheduling model is built; S3, the improved quantum genetic algorithm is adopted for carrying out optimal solution on the multi-target economic scheduling model, and the optimal active power of each micro source is acquired; and S4, according to the optimal active power of each micro source, active power output by each micro source is controlled. Compared with the prior art, the micro-grid formed by distributed power sources operates in a more economic, flexible and environment-friendly mode, and power generation advantages of the distributed power sources can be taken.

Description

Microgrid economic optimization traffic control method based on modified model quantum genetic algorithm

Technical field

The present invention relates to micro-capacitance sensor field, especially relate to a kind of microgrid economic optimization traffic control method based on modified model quantum genetic algorithm.

Background technology

Micro-capacitance sensor (micro-grid) is also translated into microgrid, is a kind of new network structure, is the system unit of one group of micro battery, load, energy-storage system and control device composition.Micro-capacitance sensor is an autonomous system being capable of self-contr ol, protection and management, both can be incorporated into the power networks with external electrical network, it is also possible to isolated operation.Micro-capacitance sensor is a concept of relatively conventional bulk power grid, refers to multiple distributed power source and related load thereof the network according to certain topological structure composition, and is coupled to normal grid by static switch pass.

Exploitation and extend micro-capacitance sensor and can sufficiently promote the extensive access of distributed power source and regenerative resource, it is achieved the highly reliable supply to load various energy resources form, is a kind of effective means realizing active power distribution network is tradition electrical network to intelligent grid transition.

Along with the development of national economy, electricity needs increases rapidly, and power department concentrates on investment in the construction of the large-scale centralized power supplys such as thermoelectricity, water power and nuclear power and supertension remote conveying electrical network mostly.But, along with the continuous expansion of electrical network scale, the drawback of ultra-large power system also shows especially day by day, and cost is high, runs difficulty big, it is difficult to adapts to the increasingly higher security and the reliability of user and requires and diversified power demands.

In the last few years under the overall background that electricity changes, microgrid energy optimum management increasingly receives attention.Microgrid energy Optimized Operation, by micro-source of distributed power generation (distributedgeneration, DG), energy-storage units, load and electrical network current operating conditions and historical data are analyzed, makes assessment and the prediction of science then；Different choice according to priority scheduling power classification, load rating and major network system electricity price type that all types of distributed power sources in micro-grid system are enjoyed, different energy scheduling strategies, determine corresponding Optimal Operation Model, adopt the optimized operation plan of effective Algorithm for Solving difference in future dispatching cycle, including to the plan of exerting oneself of degree type unit adjustable in microgrid, energy-storage units operation plan a few days ago and Real-Time Scheduling plan a few days ago, electric energy and heat energy are provided to user simultaneously, realize cogeneration of heat and power (combinedheatandpower, CHP).For the load that electric automobile (electricvehicles, EV) user is microgrid, also it is can as power supply.It is significant for economy, environment and energy security problem etc. that EV rationally accesses microgrid.Therefore, the scheduling problem of micro-capacitance sensor, just it is increasingly subject to pay close attention to.

Summary of the invention

Defect that the purpose of the present invention is contemplated to overcome above-mentioned prior art to exist and a kind of microgrid economic optimization traffic control method based on modified model quantum genetic algorithm is provided, the microgrid that can realize being made up of distributed power source is with more economical, flexibly, the mode of environmental protection is run, it is possible to give full play to the generating advantage of distributed power source.

The purpose of the present invention can be achieved through the following technical solutions:

A kind of microgrid economic optimization traffic control method based on modified model quantum genetic algorithm, described microgrid is in grid-connect mode running status, including multiple micro-sources and load, each micro-source and load are connected by interconnection, described load includes electric load and thermic load, described micro-source includes miniature gas turbine (microturbine, MT), blower fan (windturbine, WT), photovoltaic cell (photovoltaiccell, PV), fuel cell (fuelcell, FC), accumulator (storagebattery, and electric automobile (EV) SB), the method comprises the following steps:

S1: obtain the status information in each load and each micro-source in microgrid；

S2: set up multiple target economic load dispatching model for target with operating cost and pollutant control cost minimization；

S3: adopt modified model quantum genetic (improvedquantumgeneticalgorithm, IQGA) algorithm, by the status information in load each in microgrid and each micro-source to multiple target economy (multi-objectiveoptimization, MO) scheduling model is optimized and solves, it is thus achieved that the optimum active power in each micro-source；

S4: according to the output of the active power in each micro-source of optimum active power controller in each micro-source obtained.

Described multiple target economic load dispatching model includes object function and constraints, and described object function Z meets below equation:

Z=min{wF₁+(1-w)F₂}

$F_1=\underset{i=1}{\overset{N}{\Σ}}{C}_i^{Install}+\underset{t=1}{\overset{T}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}(C_{i,t}^{Om}+C_{i,t}^{Fuel})+\underset{t=1}{\overset{T}{\Σ}}{C}_t^{Grid}$

$F_2=\underset{t=1}{\overset{T}{\Σ}}\left\{\underset{j=1}{\overset{W}{\Σ}}{\mathrm{\μ}}_j(\underset{i=1}{\overset{N}{\Σ}}{K}_{ij}{P}_{i,t}+K_{Grid,j}{P}_{Grid,t})\right\}$

In formula: F₁For operating cost, F₂For pollutant control cost, w is weight coefficient, 0≤w≤1, and N is micro-source installation sum, and T is the total duration of simulation optimization,It is i-th kind of micro-source mounting cost,For i-th kind of micro-source operation and maintenance cost in the t time period,For i-th kind of fuel used expense in micro-source in the t time period,For exchanging the expense of power in the t time period between microgrid and outer net, W is pollutant sums, μ_jFor jth pollutant handling use, K_ijIt is i-th kind of micro-source jth pollutant emission factor, P_i,tFor the active power of i-th kind of micro-source output, K in the t time period_Grid,jFor outer net jth pollutant emission factor, P_Grid,tFor microgrid in the t time period and the mutual electricity of outer net；

Described constraints includes equality constraint and inequality constraints condition, described equality constraint includes active power balance constraint and discharge and recharge of accumulator constraint, and described inequality constraints condition includes accumulator and runs constraint, electric automobile discharge and recharge constraint, microgrid permission and the mutual power constraint of outer net and miniature gas turbine, blower fan, the constraint of photovoltaic cell active power of output.

Described i-th kind of micro-source mounting costMeet below equation:

C_i ^Install=R_i ^Install×S_i, i=1 ..., N

In formula: R_i ^InstallIt is the unit installation cost in i-th kind of micro-source, S_iIt it is the initial installed capacity in i-th kind of micro-source；

I-th kind of micro-source operation and maintenance cost in the described t time periodMeet below equation:

$C_{i,t}^{\mathrm{Om}}={R_{i,t}}^{\mathrm{Om}}\×P_{i,t},i=1,...,N,t=1,...,T$

In formula: R_i,t ^OmFor the unit operation and maintenance cost in i-th kind of micro-source in the t time period；

I-th kind of fuel used expense in micro-source in the described t time periodMeet below equation:

C_i,t ^Fuel=R_i,t ^Fuel×P_i,t, i=1 ..., N, t=1 ..., T

In formula: R_i,t ^FuelFor the unit of fuel expense in i-th kind of micro-source in the t time period；

The expense of power is exchanged between microgrid and outer net in the described t time periodMeet below equation:

C_t ^Grid=R_Grid,t×P_Grid,t, t=1 ..., T

In formula: R_Grid,tSpot Price for t time period intranet and extranet.

Described active power balance retrains:

$\underset{i}{\overset{N}{\mathrm{\Σ}}}{P}_{i,t}+P_{\mathrm{SB},t}+P_{\mathrm{Grid},t}-P_{\mathrm{Load},t}=0,t=1,...,T$

In formula: P_SB,tCharge capacity, P is put for accumulator in the t time period_Load,tPower is consumed for t time period internal loading；

The discharge and recharge of described accumulator retrains:

SOC_t-P_Batt,t/BAT_cap=SOC_t+1, t=1 ..., T

In formula: SOC_tFor the SOC value of accumulator, P in the t time period_Batt,tFor the active power of accumulator output, BAT in the t time period_capFor accumulator total capacity.

Described accumulator runs constraint:

$P_{\mathrm{SB}}^{\min }\≤P_{\mathrm{SB},t}\≤P_{\mathrm{SB}}^{\max },t=1,...,T$

-S_inv,SB≤P_SB,t≤S_inv,SB, t=1 ..., T

In formula:For the minimum active power of accumulator,For the maximum active power of accumulator, P_SB,tFor the active power of accumulator output, S in the t time period_inv,SBFor the inverter capacity that accumulator is corresponding；

Described electric automobile discharge and recharge retrains:

$P_{EV}^{\min }\≤P_{EV,t}\≤P_{EV}^{\max },t=1,...,T$

-S_inv,EV≤P_EV,t≤S_inv,EV, t=1 ..., T

In formula:For the minimum active power of electric automobile,For the maximum active power of electric automobile, P_EV,tFor the active power of electric automobile output, S in the t time period_inv,EVFor the inverter capacity that electric automobile is corresponding；

Described microgrid allows and the mutual power constraint of outer net:

$P_{Grid,t}^{\min }\≤P_{Grid,t}\≤P_{Grid,t}^{\max },t=1,...,T$

In formula:For microgrid in the t time period and the mutual power upper limit of outer net, lower limit；

Described miniature gas turbine, blower fan, photovoltaic cell active power of output retrain:

$P_{k,t}^{\min }\≤P_{k,t}\≤P_{k,t}^{\max },t=1,...,T$

In formula:The respectively upper limit of active power of kth kind micro-source output, lower limit, P in miniature gas turbine, blower fan, photovoltaic cell in the t time period_k,tActive power for the micro-source output of kth kind in miniature gas turbine, blower fan, photovoltaic cell in the t time period.

Described modified model quantum genetic algorithm refers to: in quantum genetic algorithm, the coding of population adopts double chain framework, simultaneously in Evolution of Population process, adopts dynamic quantum door rotation angle degree to carry out Population Regeneration.

Compared with prior art, the invention have the advantages that

1) problem that the present invention is directed to microgrid Optimized Operation, propose to take into account, with electric automobile, the microgrid multiple target economic load dispatching model heating income simultaneously as load and generator unit simultaneously, and build energy the Internet and determine constraints, with maximization of economic benefit, Environmental costs minimize as microgrid multi-objective optimization question, real power control can be optimized, can measure with meritorious by Integrated comparative load, choose optimal scheduling scheme, load condition according to different periods, micro-capacitance sensor can obtain reasonable arrangement, and can be scheduling as required, load or excision load is accessed in time according to the situation that system access and output are meritorious, reach the purpose of wastage reducing and energy saving, the microgrid realizing being made up of distributed power source is with more economical, flexibly, the mode of environmental protection is run, the generating advantage of distributed power source can be given full play to.

2) present invention introduces double chain framework and dynamic rotating angle, propose a kind of new modified model quantum genetic algorithm to solve, show that micro-capacitance sensor optimizes traffic control scheme, through simulation results show: Improving Genetic Algorithm compares common genetic algorithm and basis quantum genetic algorithm, has better global optimizing ability, convergence rate and robustness.

3) adopt the inventive method can conveniently carry out each micro-source active power controller in microgrid, greatly reduce the workload that operations staff's scheduler routine calculates.

Accompanying drawing explanation

Fig. 1 is the inventive method flow chart；

Fig. 2 is IQGA algorithm flow chart in the present invention；

Fig. 3 is microgrid internal structure schematic diagram in the present invention；

Fig. 4 is the consumption power schematic diagram of thermic load and electric load in microgrid in embodiment；

Fig. 5 is the output power curve schematic diagram of photovoltaic cell and blower fan in embodiment；

Fig. 6 is the power curve schematic diagram of accumulator in embodiment, miniature gas turbine, fuel cell and outer net power；

Fig. 7 is accumulator SOC curve synoptic diagram in embodiment；

Fig. 8 is containing electric automobile and battery mixed energy storage system power curve schematic diagram in embodiment；

Fig. 9 is microgrid optimum results schematic diagram under grid-connect mode in embodiment；

Figure 10 is microgrid scheduling result schematic diagram under grid-connect mode in embodiment；

Figure 11 is that in embodiment, under grid-connect mode, exert oneself schematic diagram in each micro-source of microgrid；

Figure 12 is modified model quantum genetic algorithm in embodiment, common genetic algorithm and basis quantum genetic algorithm evolutionary process comparison diagram.

Detailed description of the invention

Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.The present embodiment is carried out premised on technical solution of the present invention, gives detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.

As shown in Figure 1, a kind of microgrid economic optimization traffic control method based on modified model quantum genetic algorithm, microgrid is in grid-connect mode running status, including multiple micro-sources and load, each micro-source and load are connected by interconnection, load includes electric load and thermic load, micro-source includes miniature gas turbine, blower fan, photovoltaic cell, fuel cell, accumulator and electric automobile, wherein miniature gas turbine, blower fan, photovoltaic cell is distributed generation unit, fuel cell and accumulator are hybrid battery energy-storage units, electric automobile may act as load can also as distributed generation unit, between microgrid and outer net (Grid), trend is mutual, the method comprises the following steps:

S1: the status information in each load and each micro-source in microgrid in obtaining a day, wherein by setting up the mathematical model in each micro-source, can obtain corresponding status information, specifically have:

1, electric automobile model:

What EV ran offer energy is ultracapacitor, ignores the self-discharge rate of super capacitor, then the model of EV energy-storage units is:

$S=\frac{Q}{Q_N}=\frac{C(U_C-U_{min})}{C(U_{max}-U_{\min })}=\frac{U_0+1/C{\∫}_0^t{I}_{C}dt-U_{min}}{U_{max}-U_{min}}$

The energy stored in EV is:

$W=1/2C(U_{max}^2-U_{min}^2)$

In formula: S is the dump energy after EV discharge and recharge, Q is the actual quantity of electric charge of super capacitor storage, Q_NFor the maximum amount of charge of super capacitor storage, U_max、U_minRespectively ultracapacitor running voltage the highest, minimum, U₀For ultracapacitor initial voltage, I_CFor charging and discharging currents.

2, miniature gas turbine model:

Miniature gas turbine mathematical model is:

Q_MT(t)=P_e(t)(1-η_e(t)-η₁)/η_e(t)

In formula: Q_MTT () is t gas turbine waste heat amount, P_eT electrical power that () exports for t gas turbine, η_eT () is the generating efficiency of t gas turbine, η₁Radiation loss coefficient for t gas turbine.

Q_he(t)=Q_MT(t)K_he

In formula: Q_heT heating capacity that () supplies for t gas turbine flue gas waste heat, K_heFor bromine cooling machine heating efficiency.

V_MT=Σ (P_e(t)Δt/(η_e(t)L))

In formula: V_MTFor the amount of natural gas that gas turbine consumes, t is the operation time of gas turbine, and L is natural gas low heat value, is taken as 9.7kW h/m³。

The fuel cost calculation of miniature gas turbine is:

C_MT=(C_n1/L)Σ(P_e(t)Δt/(η_e(t))

In formula: C_MTFor gas turbine fuel cost, C_n1For Gas Prices, take 0.5 $ (unit)/m³。

nullPhotovoltaic cell model is shown in that document " Study on economical operation during micro-grid connection " (engrave by cattle，Huang Wei，Guo Jiahuan，Deng. electric power network technique，2010，34(11):38-42)，Document " SystemmodelingandoptimizationofmicrogridusingGeneticAlgo rithm " (QijunDeng is shown in by blower fan model,XingGao,HongZhou.The2ndInternationalConferenceonIntelligentControlandInformationProcessing,2011:540-544)，Miniature gas turbine model、Fuel cell and battery model can referring to document " Study on economical operation of microgrid " (Li Le. Beijing: North China Electric Power University，2011,3).

S2: with operating cost and pollutant control cost minimization be target, build containing electric automobile EV, photovoltaic cell PV, blower fan WT, miniature gas turbine MT, fuel cell FC, accumulator SB and other types distributed power source and comprise thermoelectricity load simultaneously set up multiple target economic load dispatching model, considering that electric automobile is simultaneously as load and generator unit, and take into account cogeneration of heat and power and heat on the basis of income, build energy the Internet.

1) microgrid operating cost is considered

The operating cost F of microgrid₁Including installation cost, operation expense, fuel cost and outer net purchase sale of electricity cost, with F₁The minimum object function Z for target₁For:

Z₁=minF₁

$F_1=\underset{i=1}{\overset{N}{\Σ}}{C}_i^{Install}+\underset{t=1}{\overset{T}{\Σ}}\underset{i=1}{\overset{N}{\Σ}}(C_{i,t}^{Om}+C_{i,t}^{Fuel})+\underset{t=1}{\overset{T}{\Σ}}{C}_t^{Grid}$

In formula: N is micro-source installation sum, and T is the total duration of simulation optimization,It is i-th kind of micro-source mounting cost,For i-th kind of micro-source operation and maintenance cost in the t time period,For i-th kind of fuel used expense in micro-source in the t time period,For exchanging the expense of power in the t time period between microgrid and outer net.

(1) i-th kind of micro-source mounting costMeet below equation:

C_i ^Install=R_i ^Install×S_i, i=1 ..., N

In formula: R_i ^InstallIt is the unit installation cost (unit/kW) in i-th kind of micro-source, S_iIt it is the initial installed capacity (kW) in i-th kind of micro-source；

(2) i-th kind of micro-source operation and maintenance cost in the t time periodMeet below equation:

$C_{i,t}^{\mathrm{Om}}={R_{i,t}}^{\mathrm{Om}}\×P_{i,t},i=1,...,N,t=1,...,T$

In formula: R_i,t ^OmFor the unit operation and maintenance cost [unit/(kW h)] in i-th kind of micro-source, P in the t time period_i,tFor the active power of i-th kind of micro-source output, P in the t time period_i,tCorresponding distributed generation unit (miniature gas turbine, blower fan, photovoltaic cell) is unit time (1h) generated energy (active power, kW), P_i,tCorresponding hybrid battery energy-storage units or electric automobile are unit time charge-discharge electric power (active power, kW)；

(3) i-th kind of fuel used expense in micro-source in the t time periodMeet below equation:

C_i,t ^Fuel=R_i,t ^Fuel×P_i,t, i=1 ..., N, t=1 ..., T

In formula: R_i,t ^FuelFor the unit of fuel expense [unit/(kW h)] in i-th kind of micro-source in the t time period, regenerative resource distributed generation unit is owing to need not use fuel, and this takes 0；

(4) expense of power is exchanged in the t time period between microgrid and outer netMeet below equation:

C_t ^Grid=R_Grid,t×P_Grid,t, t=1 ..., T

In formula: R_Grid,tFor the Spot Price of t time period intranet and extranet, P_Grid,tFor microgrid in the t time period and the mutual electricity of outer net, P_Grid,tTake sign situation: just take when the outside net purchase electricity of microgrid, take when microgrid is to outer net sale of electricity negative.

2) pollutant control cost is considered

With pollutant control cost F₂The minimum object function Z for target₂For:

Z₂=minF₂

$F_2=\underset{t=1}{\overset{T}{\Σ}}\left\{\underset{j=1}{\overset{W}{\Σ}}{\mathrm{\μ}}_j(\underset{i=1}{\overset{N}{\Σ}}{K}_{ij}{P}_{i,t}+K_{Grid,j}{P}_{Grid,t})\right\}$

In formula: W is pollutant (CO₂、SO₂、NO_xDeng) sum, μ_jFor jth pollutant handling use, K_ijIt is i-th kind of micro-source jth pollutant emission factor, K_Grid,jFor outer net jth pollutant emission factor, regenerative resource distributed generation unit is absent from pollutant emission, and this takes 0.

3) economy and environmental benefit are considered

By operating cost F₁With pollutant control cost F₂It is weighted, can F₁、F₂Single-objective problem is converted into multi-objective optimization question and solves, the multiple target economic load dispatching model then set up includes object function and constraints, object function Z be microgrid in one day cost of electricity-generating minimum, microgrid in one day cost of electricity-generating include microgrid operating cost and pollutant control cost, meet below equation:

Z=min{wF₁+(1-w)F₂}

In formula: w is weight coefficient, 0≤w≤1, it is possible to specifically arrange according to microgrid and be adjusted.

Constraints includes equality constraint and inequality constraints condition, and equality constraint includes active power balance constraint and discharge and recharge of accumulator constraint, specific as follows:

Active power balance retrains:

$\underset{i}{\overset{N}{\Σ}}{P}_{i,t}+P_{SB,t}+P_{Grid,t}-P_{Load,t}=0,t=1,...,T$

In formula: P_SB,tPutting charge capacity for accumulator in the t time period, electric discharge just takes, and charging takes negative, P_Load,tPower is consumed for t time period internal loading；

Discharge and recharge of accumulator retrains:

SOC_t-P_Batt,t/BAT_cap=SOC_t+1, t=1 ..., T

In formula: SOC_tFor SOC (StateofCharge, the dump energy) value of accumulator, P in the t time period_Batt,tFor the active power of accumulator output, BAT in the t time period_capFor accumulator total capacity.

Inequality constraints condition includes accumulator and runs constraint, electric automobile discharge and recharge constraint, microgrid permission and the mutual power constraint of outer net and miniature gas turbine, blower fan, the constraint of photovoltaic cell active power of output, specific as follows:

Accumulator runs constraint:

$P_{SB}^{\min }\≤P_{SB,t}\≤P_{SB}^{\max },t=1,\mathrm{...},T$

-S_inv,SB≤P_SB,t≤S_inv,SB, t=1 ..., T

In formula:For the minimum active power of accumulator,For the maximum active power of accumulator, P_SB,tFor the active power of accumulator output, S in the t time period_inv,SBFor the inverter capacity that accumulator is corresponding；

Electric automobile discharge and recharge retrains:

$P_{EV}^{\min }\≤P_{EV,t}\≤P_{EV}^{\max },t=1,...,T$

-S_inv,EV≤P_EV,t≤S_inv,EV, t=1 ..., T

In formula:For the minimum active power of electric automobile,For the maximum active power of electric automobile, P_EV,tFor the active power of electric automobile output, S in the t time period_inv,EVFor the inverter capacity that electric automobile is corresponding；

Microgrid allows and the mutual power constraint of outer net:

$P_{Grid,t}^{\min }\≤P_{Grid,t}\≤P_{Grid,t}^{\max },t=1,...,T$

In formula:For microgrid in the t time period and the mutual power upper limit of outer net, lower limit；

Miniature gas turbine, blower fan, photovoltaic cell active power of output retrain:

$P_{k,t}^{\min }\≤P_{k,t}\≤P_{k,t}^{\max },t=1,...,T$

In formula:The respectively upper limit of active power of kth kind micro-source output, lower limit, P in miniature gas turbine, blower fan, photovoltaic cell in the t time period_k,tActive power for the micro-source output of kth kind in miniature gas turbine, blower fan, photovoltaic cell in the t time period.

S3: adopt modified model quantum genetic algorithm, is optimized multiple target economic load dispatching model by the status information in load each in microgrid and each micro-source and solves, it is thus achieved that Optimized Operation scheme in a day 24 hours, including the optimum active power in each micro-source.

Considering that micro-capacitance sensor economic optimization operation problem is a complicated nonlinear problem, the present invention introduces double chain framework and dynamic rotating angle on classical genetic algorithm basis, is specifically described below:

Wherein double chain framework adopts the binary coding in genetic algorithm, carrying out quantum bit coding to there is polymorphic problem, adopting four states to be encoded with two quantum bits.The gene adopting muliti-qubit m parameter of coding is as follows:

$q_j^t=\[\left.\begin{array}{c}{\mathrm{\α}}_{11}^t\\ {\mathrm{\β}}_{11}^t\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{12}^t\\ {\mathrm{\β}}_{12}^t\end{array}\right|\left.\begin{array}{c}\mathrm{...}\\ \mathrm{...}\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{1k}^t\\ {\mathrm{\β}}_{1k}^t\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{21}^t\\ {\mathrm{\β}}_{21}^t\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{22}^t\\ {\mathrm{\β}}_{22}^t\end{array}\right|\left.\begin{array}{c}\mathrm{...}\\ \mathrm{...}\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{2k}^t\\ {\mathrm{\β}}_{2k}^t\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{m1}^t\\ {\mathrm{\β}}_{m1}^t\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_{m2}^t\\ {\mathrm{\β}}_{m2}^t\end{array}\right|\left.\begin{array}{c}\mathrm{...}\\ \mathrm{...}\end{array}\right|\begin{array}{c}{\mathrm{\α}}_{mk}^t\\ {\mathrm{\β}}_{mk}^t\end{array}\]$

In formula:Representing the t chromosome for jth individuality, k is the quantum bit number encoding each gene；M is chromosomal gene number.

In double chain framework, population isWherein n is scale,Representing a quantum chromosomes, m is quantum bit number, it is contemplated that randomness and the quantum state probability amplitude of initialization of population should meet normalized binding character, and the chromosomal double-strand coding of i-th is defined as:

$q_i^t=\[\left|\begin{array}{c}\cos \left(t_{i1}\right)\\ \sin \left(t_{i1}\right)\end{array}\right.\left|\begin{array}{c}\cos \left(t_{i2}\right)\\ \sin \left(t_{i2}\right)\end{array}\right.\left|\begin{array}{c}\mathrm{...}\\ \mathrm{...}\end{array}\right.\left|\begin{array}{c}\cos \left(t_{ij}\right)\\ \sin \left(t_{ij}\right)\end{array}\right.\left|\begin{array}{c}\mathrm{...}\\ \mathrm{...}\end{array}\right.\left|\begin{array}{c}\cos \left(t_{im}\right)\\ \sin \left(t_{im}\right)\end{array}\right.\]$

Wherein, t_ij=2 π × rand, rand is the random number between (0,1), i=1,2 ..., n, j=1,2 ..., m.Every chromosome comprises two gene strands arranged side by side, and every gene strand can represent an optimization solution.Therefore, every chromosome represents two optimal solutions in search volume:

P_ic=(cos (t_i1),cos(t_i2),...,cos(t_im))

P_is=(sin (t_i1),sin(t_i2),...,sin(t_im))

In formula: P_icIt is called that " cosine " solves, P_isIt is called that " sine " solves.

Basis quantum genetic algorithm uses fixing anglec of rotation strategy, and the IQGA that the present invention proposes can according to the anglec of rotation size of the dynamic adjustment amount cervical orifice of uterus of evolution process.Algorithm initial operating stage arranges the bigger anglec of rotation, along with the increase of evolutionary generation is gradually reduced the anglec of rotation.Adjustable strategies is to individualityMeasurement, assesses its fitness f (x_j)^t, compare with the fitness value f (best) of the optimum individual retained, according to comparative result adjustmentMiddle corresponding positions quantum bit so that (α, β) evolve in the direction of solution towards being conducive to optimum to determine.

s(α_i,β_i) dynamic rotary angle selection strategy is as shown in table 1.In table 1, x_iFor the i-th bit of current chromosome, best_iFor the current chromosomal i-th bit of optimum, x '_iFor changing the i-th bit best of after stain colour solid_iFor the current chromosomal i-th bit of optimum, f (x) is fitness function, s (α_i,β_i) for anglec of rotation direction, Δ θ_iFor present rotation angel degree size, Δ θ '_iFor anglec of rotation size after changing, its value is determined by selection strategy listed in table 1.

Table 1 dynamic rotary angle selection strategy

In table, the expression formula of γ is:

γ=0.002 π+0.004 π * ((b.fitness-fitness (i))/b.fitness

+0.5*exp(1)^(1-MAXGEN/gen)

In formula: b.fitness is adaptive optimal control angle value, fitness (i) is current fitness value, and MAXGEN is maximum evolutionary generation, and gen is current evolutionary generation.

This adjustable strategies is by individualityFitness value f (the best of fitness f (x) of current measured value and the currently most individuality of this population_i) compare, if f (x) > f is (best_i), then adjustMiddle corresponding positions quantum bit so that probability amplitude is to (α_i,β_i) towards being conducive to x_iThe direction occurred develops；Otherwise, if f (x) < f is (best_i), then adjustMiddle corresponding positions quantum bit so that probability amplitude is to (a_i,β_i) towards being conducive to the best direction occurred to develop.

As in figure 2 it is shown, the detailed process of IQGA describes as follows:

(1) population Q (t is initialized₀), stochastic generation digs the individual chromosome with quantum bit for coding, and subscript 0 represents initial value；

(2) to initial population Q (t₀) in each individuality carry out one-shot measurement, obtain correspondence determination solution P (t₀)；

(3) determine that solution carries out Fitness analysis to each；

(4) record optimum individual and corresponding fitness；

(5) judging whether calculating process can terminate, if meeting termination condition, exiting, otherwise continue to calculate；

(6) judging whether to adopt new quantum door rotation angle degree, if not adopting, continuing to calculate；According to then jumping to step (12)；

(7) to each individual enforcement one-shot measurement in population Q (t), solution is determined accordingly；

(8) determine that solution carries out Fitness analysis to each；

(9) utilize Quantum rotating gate U (t) to implement to adjust to individuality, obtain new population Q (t+1)；

(10) record optimum individual and corresponding fitness；

(11) iterations t is added 1, return step (5)；

(12) new quantum door rotation angle degree is calculated；

(13) to each individual enforcement one-shot measurement in population Q (t), solution is determined accordingly；

(14) determine that solution carries out Fitness analysis to each；

(15) utilize Quantum rotating gate U (t) to implement to adjust to individuality, obtain new population Q (t+1)；

(16) record optimum individual and corresponding fitness；

(17) iterations t is added 1, return step (5).

S4: according to the output of the active power in each micro-source of optimum active power controller in each micro-source obtained.

The microgrid structure adopted in the present embodiment is as shown in Figure 3.Microgrid in Fig. 3 is made up of industrial load C, thermoelectricity load D, miniature gas turbine MT, resident load A, Commercial Load B, blower fan WT, fuel cell FC, photovoltaic cell PV, electric automobile EV, permission interruptible load E and accumulator SB etc., 1,2,3,7,8,13,15 represent wiring point, PCC represents the points of common connection of microgrid and outer net, and microgrid is in grid-connect mode running status.Battery rating is 800kW h, and depth of discharge is charged as the best when being 50%-75%, and therefore initial quantity of electricity is set to 60%, and efficiency for charge-discharge is 1, ignores self discharge, and the inverter capacity of accumulator is 50kVA；The inverter capacity of fuel cell is 30kVA；The power upper and lower limit of electric automobile is set to 30kW ,-30kW, and its inverter capacity is 30kVA；Operation of fuel cells in electricity determining by heat, therefore it to send power with miniature gas turbine linear, heat income be 0.12 yuan/(kW h)；When being 8 during electricity consumption peak～20 time, when being 20 during electricity consumption paddy～8 time.

The Sino-Japan electric load of microgrid and day thermic load situation as shown in Figure 4；Each distributed electrical source dates is in Table 2；The installation cost in each micro-source and energy cost are in Table 3；The active power curves of intermittence generating micro-source PV, WT is as shown in Figure 5；The emissions data of each micro-source pollutants is in Table 4.

The basic parameter in each micro-source of table 2

The installation cost in each micro-source of table 3 and energy cost

The emissions data of each micro-source pollutants of table 4

With Financial cost and Environmental costs for target, above-mentioned microgrid is optimized.In microgrid, thermic load-electric power curves is as shown in Figure 6, and accumulator SOC is as shown in Figure 7.From Fig. 6, Fig. 7, when 0～8 time microgrid load also relatively lighter, remaining electric energy is charged by distributed power source to accumulator SB；When 8～20 time, microgrid exists meritorious vacancy, PV, WT and MT the workload demand that can not meet in microgrid of exerting oneself, at this moment call SB electric discharge, it is ensured that micro net power quality；When 8～15 time between, SB can meet the workload demand in microgrid, it is possible to reduce calls the meritorious of FC and exerts oneself.When 20～6 time, SB is charged, owing to the Environmental costs of FC are relatively low, preferentially call FC provide active power.FC works in electricity determining by heat pattern, therefore sends power with WT linear.

Containing electric automobile EV and battery mixed energy storage system power curve as shown in Figure 8.As seen from Figure 8, when SB, FC and EV are collectively as back-up source, battery system and both EV cooperate and make the power in microgrid more mild；And when meritorious vacancy occurs in microgrid, decrease microgrid purchase of electricity from outer net, improve the economic benefit of microgrid.Meanwhile, EV to microgrid sale of electricity, can be charged when power abundance in microgrid, well serve the effect of " peak load shifting " when meritorious shortage of power occurs in microgrid.

Then under grid-connect mode microgrid optimum results as it is shown in figure 9, under grid-connect mode microgrid scheduling result is as shown in Figure 10.From Fig. 9, Figure 10, PV, WT, SB and EV energy cost be relatively low, microgrid should preferentially call free of contamination PV and WT and meet power demand；Owing to the energy cost of FC is apparently higher than SB, therefore should preferentially calling SB, EV user to microgrid transmission power, so can reduce the discharge and recharge number of times of SB simultaneously, extends the SB life-span, can recall FC when meritorious vacancy occurs in microgrid；WT cost is the highest, environmental pollution is maximum, should finally consider to call.

The optimum results that under grid-connect mode, in microgrid, exert oneself in each micro-source is as shown in figure 11.As seen from Figure 11, PV and WT works in MPPT maximum power point tracking pattern；Charging the best owing to SB depth of discharge is 50%～75%, from 7 time～18 time before the discharge mode of SB be maintained at 75%；When 6～18 time, EV user carries out selling electricity to microgrid according to microgrid load condition；When 0～7 time and when 18～24 time, in microgrid, load is relatively light, and PV and WT charges to SB after meeting optimized distributionl requirement, and SB is in charge mode；Now preferentially call FC, recall MT and meet meritorious vacancy.When 7～18 time, in microgrid, power consumption is substantially increased, and SB is in discharge mode, and EV user also carries electric energy to microgrid, reduces the exerting oneself of FC and MT.

In order to the advantage of the IQGA algorithm adopted herein is described, below it is compared with traditional GA algorithm and basis QGA algorithm.

Simulation parameter is as follows, and the population scale of modified model quantum genetic algorithm, common genetic algorithm and basis quantum genetic algorithm is 100, and maximum genetic algebra is 200, and the binary length of each variable is 20, and weight coefficient w is taken as 0.8.As shown in figure 12, simulation architecture is as shown in table 5 for the evolution curve of three kinds of algorithms.

Table 5 simulation result

By simulation optimization result it can be seen that adopt the inventive method that microgrid is controlled, it may be achieved following scheduling strategy:

(1) preferentially utilize the clean energy resource such as the wind-powered electricity generation WT within microgrid, photovoltaic PV generating to meet workload demand and free Power Exchange can be carried out with major network；

(2) WT and PV power generation operation is in MPPT maximum power point tracking pattern；

(3) MT works in the electricity determining by heat method of operation, thermic load determine that the meritorious of MT is exerted oneself；

(4) when WT, PV and MT meritorious exert oneself satisfied whole electric load time, first electric automobile EV and accumulator SB charging is given, monitor the charging and discharging state of accumulator simultaneously, it is charged to and then fuel cell FC, can consider, when EV is full of, WT or PV that cut-out cost of electricity-generating is higher successively；

(5) EV improves microgrid benefit according to by different periods charge and discharge control, arbitrarily can charge to EV when microgrid electricity abundance；When micro-grid power source deficiency is to major network power purchase, considering for economy and stability, EV does not allow charging, and remaining electricity can be sold microgrid by EV；

(6) when meritorious the exerting oneself of WT, PV and MT cannot meet microgrid so during load, prioritizing selection accumulator is discharged, then recalling FC carry out active power of output as still suffered from meritorious vacancy, dump energy in car can be sold microgrid thus obtaining income by EV user during this period；

(7) if all micro-sources still can not meet microgrid safe and reliable operation in the scope of exerting oneself, then excise successively according to the significance level of load.

Employing the inventive method has the advantage that

(1) whole economic efficiency making micro-capacitance sensor is better；

(2) micro-capacitance sensor environmental situation is improved；

(3) service efficiency in each micro-source in micro-capacitance sensor is improved；

(4) electric automobile is made full use of thus playing the effect of " peak load shifting "；

(5) service life of accumulator and fuel cell is extended；

(6) cogeneration of heat and power operational mode improves economy and the feature of environmental protection of micro-capacitance sensor；

(7) introducing double chain framework and dynamic rotating angle, Improving Genetic Algorithm has better global optimizing ability, convergence rate and robustness.

Claims (5)

1. the microgrid economic optimization traffic control method based on modified model quantum genetic algorithm, described microgrid is in grid-connect mode running status, including multiple micro-sources and load, each micro-source and load are connected by interconnection, described load includes electric load and thermic load, it is characterized in that, described micro-source includes miniature gas turbine, blower fan, photovoltaic cell, fuel cell, accumulator and electric automobile, and the method comprises the following steps:

S1: obtain the status information in each load and each micro-source in microgrid；

S2: set up multiple target economic load dispatching model for target with operating cost and pollutant control cost minimization；

S3: adopt modified model quantum genetic algorithm, is optimized multiple target economic load dispatching model by the status information in load each in microgrid and each micro-source and solves, it is thus achieved that the optimum active power in each micro-source；

S4: according to the output of the active power in each micro-source of optimum active power controller in each micro-source obtained；

Described multiple target economic load dispatching model includes object function and constraints, and described object function Z meets below equation:

Z=min{wF₁+(1-w)F₂}

$F_1=\underset{i=1}{\overset{N}{\&Sigma;}}{C}_i^{Install}+\underset{t=1}{\overset{T}{\&Sigma;}}\underset{i=1}{\overset{N}{\&Sigma;}}(C_{i,t}^{Om}+C_{i,t}^{Fuel})+\underset{t=1}{\overset{T}{\&Sigma;}}{C}_t^{Grid}$

$F_2=\underset{t=1}{\overset{T}{\mathrm{\&Sigma;}}}\left\{\underset{j=1}{\overset{W}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_j\left(\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{K}_{ij}{P}_{i,t}+K_{Grid,j}{P}_{Grid,t}\right)\right\}$

In formula: F₁For operating cost, F₂For pollutant control cost, w is weight coefficient, 0≤w≤1, and N is micro-source installation sum, and T is the total duration of simulation optimization,It is i-th kind of micro-source mounting cost,For i-th kind of micro-source operation and maintenance cost in the t time period,For i-th kind of fuel used expense in micro-source in the t time period,For exchanging the expense of power in the t time period between microgrid and outer net, W is pollutant sums, μ_jFor jth pollutant handling use, K_ijIt is i-th kind of micro-source jth pollutant emission factor, P_i,tFor the active power of i-th kind of micro-source output, K in the t time period_Grid,jFor outer net jth pollutant emission factor, P_Grid,tFor microgrid in the t time period and the mutual electricity of outer net；

Described constraints includes equality constraint and inequality constraints condition, described equality constraint includes active power balance constraint and discharge and recharge of accumulator constraint, and described inequality constraints condition includes accumulator and runs constraint, electric automobile discharge and recharge constraint, microgrid permission and the mutual power constraint of outer net and miniature gas turbine, blower fan, the constraint of photovoltaic cell active power of output.

2. the microgrid economic optimization traffic control method based on modified model quantum genetic algorithm according to claim 1, it is characterised in that described i-th kind of micro-source mounting costMeet below equation:

C_i ^Install=R_i ^Install×S_i, i=1 ..., N

In formula: R_i ^InstallIt is the unit installation cost in i-th kind of micro-source, S_iIt it is the initial installed capacity in i-th kind of micro-source；

I-th kind of micro-source operation and maintenance cost in the described t time periodMeet below equation:

$C_{i,t}^{Om}={R_{i,t}}^{Om}\&times;P_{i,t},i=1,...,N,t=1,...,T$

In formula: R_i,t ^OmFor the unit operation and maintenance cost in i-th kind of micro-source in the t time period；

I-th kind of fuel used expense in micro-source in the described t time periodMeet below equation:

C_i,t ^Fuel=R_i,t ^Fuel×P_i,t, i=1 ..., N, t=1 ..., T

In formula: R_i,t ^FuelFor the unit of fuel expense in i-th kind of micro-source in the t time period；

The expense of power is exchanged between microgrid and outer net in the described t time periodMeet below equation:

C_t ^Grid=R_Grid,t×P_Grid,t, t=1 ..., T

In formula: R_Grid,tSpot Price for t time period intranet and extranet.

3. the microgrid economic optimization traffic control method based on modified model quantum genetic algorithm according to claim 1, it is characterised in that described active power balance retrains:

$\underset{i}{\overset{N}{\&Sigma;}}{P}_{i,t}+P_{SB,t}+P_{Grid,t}-P_{Load,t}=0,t=1,...,T$

In formula: P_SB,tCharge capacity, P is put for accumulator in the t time period_Load,tPower is consumed for t time period internal loading；

The discharge and recharge of described accumulator retrains:

SOC_t-P_Batt,t/BAT_cap=SOC_T+1, t=1 ..., T

In formula: SOC_tFor the SOC value of accumulator, P in the t time period_Batt,tFor the active power of accumulator output, BAT in the t time period_capFor accumulator total capacity.

4. the microgrid economic optimization traffic control method based on modified model quantum genetic algorithm according to claim 1, it is characterised in that described accumulator runs constraint:

$P_{SB}^{\min }\&le;P_{SB,t}\&le;P_{SB}^{\max },t=1,...,T$

-S_inv,SB≤P_SB,t≤S_inv,SB, t=1 ..., T

In formula:For the minimum active power of accumulator,For the maximum active power of accumulator, P_SB,tFor the active power of accumulator output, S in the t time period_inv,SBFor the inverter capacity that accumulator is corresponding；

Described electric automobile discharge and recharge retrains:

$P_{EV}^{\min }\&le;P_{EV,t}\&le;P_{EV}^{\max },t=1,...,T$

-S_inv,EV≤P_EV,t≤S_inv,EV, t=1 ..., T

In formula:For the minimum active power of electric automobile,For the maximum active power of electric automobile, P_EV,tFor the active power of electric automobile output, S in the t time period_inv,EVFor the inverter capacity that electric automobile is corresponding；

Described microgrid allows and the mutual power constraint of outer net:

$P_{Grid,t}^{\min }\&le;P_{Grid,t}\&le;P_{Grid,t}^{\max },t=1,...,T$

In formula:For microgrid in the t time period and the mutual power upper limit of outer net, lower limit；

Described miniature gas turbine, blower fan, photovoltaic cell active power of output retrain:

$P_{k,t}^{\min }\&le;P_{k,t}\&le;P_{k,t}^{\max },t=1,...,T$

In formula:The respectively upper limit of active power of kth kind micro-source output, lower limit, P in miniature gas turbine, blower fan, photovoltaic cell in the t time period_k,tActive power for the micro-source output of kth kind in miniature gas turbine, blower fan, photovoltaic cell in the t time period.

5. the microgrid economic optimization traffic control method based on modified model quantum genetic algorithm according to claim 1, it is characterized in that, described modified model quantum genetic algorithm refers to: in quantum genetic algorithm, the coding of population adopts double chain framework, simultaneously in Evolution of Population process, dynamic quantum door rotation angle degree is adopted to carry out Population Regeneration.