CN103151797A - Multi-objective dispatching model-based microgrid energy control method under grid-connected operation mode - Google Patents

Abstract

The invention relates to a multi-objective dispatching model-based microgrid energy control method under the grid-connected operation mode. A microgrid comprises a plurality of microsources and a load. The method includes the following specific steps: (1) the information of each microsource and connecting lines in the microgrid is acquired by a processor, and with minimum power generation cost and environmental cost as objectives, a multi-objective economical dispatching model is created; (2) a membership function is created to fuzzify the multi-objective economical dispatching model, so that a non-linear single-objective microgrid economical dispatching model is obtained; (3) an improved genetic algorithm is adopted to optimize the non-linear single-objective microgrid economical dispatching model, so that the optimal active and reactive power data of each microsourcre and the connecting lines are obtained, and the data are transmitted to a microgrid power controller; (4) according to the received optimal active and reactive power data, the microgrid power controller controls the active and reactive power output of each microsoruce and the connecting lines. Compared with the prior art, the method has the advantages of high control precision and the like, and can guarantee the economical operation of the microgrid.

Description

Based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model

Technical field

The present invention relates to a kind of Economical Operation of Power Systems and regulation of energy technology, especially relate to a kind of based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model.

Background technology

Along with the enhancing gradually of people's energy-conserving and environment-protective consciousness, distributed power generation has obtained increasing attention and utilization, and particularly attracts people's attention for the microgrid research that is made of the low capacity distributed power source.Microgrid combines various distributed power sources, load, energy-storage units and control device etc., forms a single controlled unit, provides simultaneously electric energy and heat energy to the user, realizes cogeneration of heat and power (combined heat and power, CHP).Microgrid has become a kind of effective technology means of the managing distributed energy in intelligent grid, has carried out broad research for the aspects such as operation control of microgrid.

The economical operation of microgrid and Optimized Operation are the focuses of studying both at home and abroad always.At present, both at home and abroad the research of microgrid economical operation mainly being simplified to single goal with a plurality of targets such as cost of electricity-generating minimum, environmental impact network minimals is optimized.But often have conflict between a plurality of targets, can cause like this optimum results of some target can not be fully up to expectations, some scholar have proposed corresponding method, has solved preferably the conversion of multi-objective problem.Domestic scholars thunder gold is brave, the Gande is strong etc. has set up the multiple target energy optimizing model of considering economy, the feature of environmental protection and reliability, adopt binary to contrast to decide the power method and determine weight coefficient, use the linear weighted function summation multi-objective optimization question is converted to single-object problem (the energy-optimised and energy-saving and emission-reduction performance analysis of distributed power generation energy supplying system, Automation of Electric Systems, 2009,33 (23): 29-36); Document " Online Management of MicroGrid with Battery Storage Using Multiobjective Optimization " (Faisal A.Mohamed, Heikki N.Koivo.POWERENG, Setubal, Portugal, 2007:231-236) set up and considered simultaneously the minimum microgrid Model for Multi-Objective Optimization with the environmental pollution minimum of operation expense, use the linear weighted function summation multi-objective problem is converted to the single goal problem, and different weight coefficients have been discussed on the impact of optimum results; Document " microgrid energy that contains distributed power generation is managed multiple-objection optimization " (Wang Xingang; Ai Qian; xuwei China etc. protecting electrical power system and control; 2009; 37 (20): 79-83) at labor on the basis of distributed power source characteristic and optimization aim Mathematical Modeling; for the optimum management that distributed power source in microgrid is exerted oneself, proposed a kind of multi-target immune algorithm of evolving based on microhabitat and optimized each little source and exert oneself.But the rack that these research institutes adopt is simpler, the active balance of general taking into account system, the impact that less consideration is idle; Also do not consider meritorious and idle the exerting oneself in the little source of Optimized Operation simultaneously, on the other hand, corresponding constraints is too simplified, and the indexs such as microgrid Tie line Power, spinning reserve, energy-storage travelling wave tube discharge and recharge and constraints are seldom considered.

Summary of the invention

Purpose of the present invention be exactly provide in order to overcome the defective that above-mentioned prior art exists a kind of control precision high, guarantee the microgrid economical operation based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model.

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

A kind of based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, described microgrid comprises a plurality of little sources and load, each little source connects by interconnection with being connected, and described load comprises electric loading and heat load, and described method specifically comprises the following steps:

1) processor obtains the information in each little source and interconnection in microgrid, sets up multiple target economic dispatch model take cost of electricity-generating and Environmental costs minimum as target;

2) set up membership function, multiple target economic dispatch model is carried out obfuscation, obtain non-linear single goal microgrid economic dispatch model;

3) adopt Improving Genetic Algorithm that non-linear single goal microgrid economic dispatch model is optimized and find the solution, obtain the meritorious and reactive power data of the optimum of each little source and interconnection, and with this transfer of data to the microgrid power controller;

4) the microgrid power controller is gained merit and the meritorious of each little source of reactive power Data Control and interconnection and reactive power output according to the optimum that receives.

Described little source comprises photovoltaic, blower fan, miniature gas turbine, fuel cell and storage battery.

Described multiple target economic dispatch model comprises target function and constraints, and described objective function F (t) is intraday cost of electricity-generating F ₁(t) minimum, the Environmental costs F of pollutant to environmental impact simultaneously ₂(t) minimum, be specially:

minF(t)＝{F ₁(t)，F ₂(t)}t＝1，2...24 (1)

F ₁(t)＝C _f(t)+C _DP(t)+C _OM(t)+C _grid(t)-C _sh(t) (2)

F ₂(t)＝C _e(t) (3)

In formula:

$C_f\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{f}_i\left(P_i\left(t\right)\right)---\left(4\right)$

$C_{\mathrm{DP}}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}[\frac{C_{\mathrm{az},i}}{{8706k}_i}\·\frac{r{(1+r)}^{n_i}}{{(1+r)}^{n_i}-1}\·P_i\left(t\right)]---\left(5\right)$

$C_{\mathrm{OM}}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{K}_{\mathrm{OM},i}{\·P}_i\left(t\right)---\left(6\right)$

C _grid(t)＝CP(t)×PR _grid(t) (7)

C _sh(t)＝Q _he(t)×K _ph (8)

$C_e\left(t\right)=\underset{i=1}{\overset{n+1}{\mathrm{\Σ}}}\left(\underset{j=1}{\overset{m}{\mathrm{\Σ}}}((V_{\mathrm{ej}}+V_j){\·Q}_{\mathrm{ij}}\·P_i\left(t\right))\right)---\left(9\right)$

C _f(t), C _DP(t), C _OM(t) be respectively t fuel cost, investment depreciable cost and the operation expense in each little source constantly; C _Grid(t) be that t moment microgrid is to the power purchase cost of outer net; C _sh(t) be the income that heats of t moment co-generation unit; C _e(t) be the constantly Environmental costs in each little source and the Environmental costs summation of the outside net purchase electricity of microgrid of t; f _iBe the fuel cost function in i little source: P _i(t) be i little source t active power output constantly; N is the number in little source, and n+1 is interconnection power; C _{Az, i}It is the installation cost as per machine capacity in i little source; k _iBe the capacity factor in i little source, k _iAnnual energy output/(rated power in 8760 * this little source) in the=the i little source; R is Annual Percentage Rate; n _iIt is the pay back period of investment in i little source; K _{OM, i}It is the unit quantity of electricity operation expense coefficient in i little source; CP (t) is that t moment microgrid is to the purchase electricity price of outer net; PR _Grid(t) be that t moment microgrid is to the purchase of electricity of outer net; Q _he(t) be the t heating capacity of microgrid constantly; K _phPrice for the system of units heat; V _ejIt is the environmental value of j item pollutant; V _jIt is the suffered fine of j item pollutant; Q _ijIt is the j item pollutant discharge amount of i little source unit quantity of electricity; M is the kind of pollutant;

Described constraints comprises:

A, trend constraint:

${\mathrm{\ΔP}}_f=P_{\mathrm{fs}}-\underset{g\∈f}{\overset{h}{\mathrm{\Σ}}}{V}_f{V}_g(G_{\mathrm{fg}}\cos {\mathrm{\θ}}_{\mathrm{fg}}+B_{\mathrm{fg}}\sin {\mathrm{\θ}}_{\mathrm{fg}})=0---\left(10\right)$

${\mathrm{\ΔQ}}_f=Q_{\mathrm{fs}}-\underset{g\∈f}{\overset{h}{\mathrm{\Σ}}}{V}_f{V}_g(G_{\mathrm{fg}}\sin {\mathrm{\θ}}_{\mathrm{fg}}+B_{\mathrm{fg}}\cos {\mathrm{\θ}}_{\mathrm{fg}})=0---\left(11\right)$

In formula: h is the nodes of microgrid, f=1, and 2 ..., h; G _fg, B _fg, θ _fgBe respectively admittance and phase angle difference between node f and node g, g ∈ f represents the node that is connected with node f:

B, working voltage constraint:

V _f，min≤V _f≤V _f，max (12)

In formula: V _{F, min}, V _{F, max}Be respectively minimum, the maximum of the upper working voltage of node f;

The active power constraint of c, photovoltaic, blower fan and miniature gas turbine:

P _i，min≤P _i≤P _i，max (13)

In formula: P _{I, min}, P _{I, max}Be respectively minimum, the maximum of i little source active power;

D, the constraint of miniature gas turbine climbing rate:

P when increasing load _MT(t)-P _MT(t-1)≤R _{Up, MT}(14)

P during load shedding _MT(t-1)-P _MT(t)≤R _{Down, MT}(15)

In formula: P _MT(t) be the t active power output of miniature gas turbine constantly, R _{Up, MT}, R _{Down, MT}Be respectively the limit value of miniature gas turbine increase and reduction active power;

E, microgrid and outer net allow mutual through-put power constraint:

0≤P _Grid(t)≤P _Grid，max (16)

0≤Q _Grid(t)≤Q _grid，max(17)

In formula: P _Grid(t), Q _Grid(t) be respectively t moment microgrid and outer net and allow mutual active power and reactive power; P _{Grid, max}, Q _{Grid, max}Be respectively and allow mutual maximum active power and reactive power;

Power factor for tie-line power transmission; C is the power factor limit value of tie-line power transmission;

F, storage battery operation constraint:

P _SB，min≤P _SB(t)≤P _SB，max (19)

-S _inv，SB≤P _SB(t)≤S _inv，SB (20)

$-\sqrt{S_{\mathrm{inv},\mathrm{SB}}^2-P_{\mathrm{SB}}^2\left(t\right)}\≤Q_{\mathrm{SB}}\left(t\right)\≤\sqrt{S_{\mathrm{inv},\mathrm{SB}}^2-P_{\mathrm{SB}}^2\left(t\right)}---\left(21\right)$

SOC _min≤SOC(t)≤SOC _max (22)

In formula: P _{SB, max}, P _{SB, min}Be respectively the minimum and maximum active power of storage battery; S _{Inv, SB}Capacity for the storage battery inverter; P _SB(t), Q _SB(t), SOC (t) be respectively t constantly the battery AC side discharge and recharge active power, reactive power and remaining battery capacity; SOC _min, SOC _maxBe respectively the minimum and maximum residual capacity of storage battery;

G, fuel cell operation constraint:

P _FC，min≤P _FC(t)≤P _FC，max (23)

P _FC(t)≤S _inv，FC (24)

$0\≤Q_{\mathrm{FC}}\left(t\right)\≤\sqrt{S_{\mathrm{mv},\mathrm{FC}}^2-P_{\mathrm{FC}}^2\left(t\right)}---\left(25\right)$

In formula: P _{FC, min}, P _{FC, max}Be respectively minimum, the maximum of fuel cell active power: P _FC(t), Q _FC(t) be respectively t active power and the reactive power of fuel cell AC constantly; S _{Inv, FC}Capacity for the fuel cell inverter;

H, spinning reserve constraint:

$P_{\mathrm{PV}}\left(t\right)+P_{\mathrm{WT}}\left(t\right)+P_{\mathrm{MT}}\left(t\right)+\min \{P_{\mathrm{SB},\max },\sqrt{{\left(S_{\mathrm{inv},\mathrm{SB}}\right)}^2-{\left(Q_{\mathrm{SB}}\left(t\right)\right)}^2},(\mathrm{SOC}(t-1)-\mathrm{SO}{C}_{\min }-$

(26)

$D_B{Q}_B^S){\mathrm{\η}}_D\}+\min \{P_{\mathrm{FC},\max },\sqrt{{\left(S_{\mathrm{inv},\mathrm{FC}}\right)}^2-{\left(Q_{\mathrm{FC}}\left(t\right)\right)}^2}\}+P_{\mathrm{Grid},\max }>=P_D\left(t\right)+P_{\mathrm{loss}}\left(t\right)+R\left(t\right)$

In formula: P _PV(t), P _WT(t) be respectively the t active power of photovoltaic, blower fan constantly; η _DBe discharging efficiency; D _BBe storage battery self discharge ratio hourly;

Total capacity for storage battery; P _D(t), P _Loss(t), R _D(t) be respectively t total load, total network loss and the required reserve capacity of micro-grid system constantly.

Described membership function adopts and falls the membership function that half Γ shape distributes, and for described target function, membership function is:

$u_k\left(t\right)=\left\{\begin{array}{cc}1& F_k\left(t\right){\≤F}_{k\min \left(t\right)}\\ \exp \left(\frac{F_{k\min }\left(t\right)-F_k\left(t\right)}{F_{k\min }\left(t\right)}\right)& F_k\left(t\right){>F}_{k\min \left(t\right)}\end{array}\right.---\left(27\right)$

In formula: F _Kmin(t) be single-goal function F _k(t) minimum value under described constraints, k=1,2; T=1,2 ..., 24.

Non-linear single goal microgrid economic dispatch model after described obfuscation is:

In formula: u (t), u ₁(t), u ₂(t) be respectively the satisfaction of fuzzy optimization, the satisfaction of cost of electricity-generating and the satisfaction of Environmental costs, wherein $u\left(t\right)=\underset{k=1}{\overset{2}{\mathrm{\∩}}}{u}_k\left(t\right).$

Described Improving Genetic Algorithm refers to: the crossover probability P in genetic algorithm _cWith the variation probability P _mCalculate according to following formula:

$P_c=\left\{\begin{array}{c}{P}_{c2}-\frac{(P_{c2}-P_{c3})(f^{\&prime;}-f_{\mathrm{avg}})}{f_{\max }-f_{\mathrm{avg}}}{f}^{\&prime;}{\&GreaterEqual;f}_{\mathrm{avg}}\\ P_{c1}-\frac{(P_{c1}-P_{c2})(f^{\&prime;}-f_{\min })}{f_{\mathrm{avg}}-f_{\min }}{f}^{\&prime;}{<f}_{\mathrm{avg}}\end{array}\right.---\left(29\right)$

$P_m=\left\{\begin{array}{c}{P}_{m2}-\frac{(P_{m2}-P_{m3})(f^{\&prime;}-f_{\mathrm{avg}})}{f_{\max }-f_{\mathrm{avg}}}{f}^{\&prime;}{\&GreaterEqual;f}_{\mathrm{avg}}\\ P_{m1}-\frac{(P_{m1}-P_{m2})(f^{\&prime;}-f_{\min })}{f_{\mathrm{avg}}-f_{\min }}{f}^{\&prime;}{<f}_{\mathrm{avg}}\end{array}\right.---\left(30\right)$

In formula: f _max, f _minAnd f _avgBe respectively maximum, minimum and average fitness in population; F ' is larger fitness in two individualities that will intersect, P _c1, P _c2, P _c3, P _m1, P _m2, P _m3Be the constant of setting, and P _c1, P _c2, P _c3∈ (0,1), P _m1, P _m2, P _m3∈ (0,0.1).

Compared with prior art, the present invention has the following advantages:

1, the present invention has considered on the basis of the economy of microgrid and green technology index, microgrid is set up multiple target economic dispatch model, more definitely reflect the actual motion state of microgrid, control precision is high, can realize environmental benefit preferably with the little operating cost of trying one's best;

2, the present invention adopts blur method and genetic algorithm that multi-objective Model is found the solution, and computational speed is fast, and solution efficiency is high;

3, the present invention is by controlling the meritorious and idle output in each little source in microgrid, itself provide idle exerting oneself to coordinate outer net idlely to exert oneself to satisfy the System Reactive Power demand to what microgrid provided by little source, the additional investment of having fallen less reactive power compensator is installed guarantees that outer net need not excessive reactive power support to be provided and to guarantee that the power that transmits between outer net and microgrid has higher power factor to microgrid.

Description of drawings

Fig. 1 is the schematic flow sheet of the inventive method;

Fig. 2 is the structural representation of micro-grid system;

Fig. 3 is to exert oneself and thermoelectricity load schematic diagram in known each little source;

Fig. 4 is microgrid active power optimum results schematic diagram under the target of cost of electricity-generating minimum;

Fig. 5 is microgrid reactive power optimum results schematic diagram under the target of cost of electricity-generating minimum;

Fig. 6 is microgrid active power optimum results schematic diagram under the minimum target of Environmental costs;

Fig. 7 is microgrid reactive power optimum results schematic diagram under the minimum target of Environmental costs;

Fig. 8 is microgrid active power optimum results schematic diagram under multiple target;

Fig. 9 is microgrid reactive power optimum results schematic diagram under multiple target;

Figure 10 is cost of electricity-generating contrast schematic diagram under single goal and multiple target;

Figure 11 is Environmental costs contrast schematic diagram under single goal and multiple target;

Figure 12 microgrid satisfaction schematic diagram under mode that is incorporated into the power networks.

Embodiment

The present invention is described in detail below in conjunction with the drawings and specific embodiments.The present embodiment is implemented as prerequisite take technical solution of the present invention, provided detailed execution mode and concrete operating process, but protection scope of the present invention is not limited to following embodiment.

Embodiment

As shown in Figure 1, a kind of based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, described microgrid comprises a plurality of little sources and load, each little source connects by interconnection with being connected, described load comprises electric loading and heat load, the microgrid of this method research is cogeneration of heat and power type micro-grid system, and the method specifically comprises the following steps:

In step 101, processor obtains the information in each little source and interconnection in microgrid, sets up multiple target economic dispatch model take cost of electricity-generating and Environmental costs minimum as target;

In step 102, set up membership function, multiple target economic dispatch model is carried out obfuscation, obtain non-linear single goal microgrid economic dispatch model;

In step 103, adopt Improving Genetic Algorithm that non-linear single goal microgrid economic dispatch model is optimized and find the solution, the optimum that obtains each little source and interconnection is gained merit and the reactive power data;

In step 104, processor with the transfer of data that obtains to the microgrid power controller;

In step 105, the microgrid power controller is gained merit and the meritorious of each little source of reactive power Data Control and interconnection and reactive power output according to the optimum that receives.

Described little source comprises (photovoltaic, PV), blower fan (wind turbine, WT), miniature gas turbine (micro turbine, MT), fuel cell (fuel cell, FC) and storage battery (storage battery, SB).In the Δ t of unit interval, this method makes the following assumptions: little source meritorious and idle exerted oneself constant; Heat, electrical load requirement is constant; Mutual power invariability between microgrid and major network and mutual electricity price are kept constant.Choose different values for Δ t, below the Optimized model set up all be suitable for, it is that 1h analyzes that the present embodiment is got Δ t.

The Mathematical Modeling in each little source is as described below:

1, miniature gas turbine model

$\left\{\begin{array}{c}{Q}_{\mathrm{MT}}\left(t\right)=P_e\left(t\right)(1-{\mathrm{\&eta;}}_e\left(t\right)-{\mathrm{\&eta;}}_1)/{\mathrm{\&eta;}}_e\left(t\right)\\ Q_{\mathrm{he}}\left(t\right)=Q_{\mathrm{MT}}\left(t\right)K_{\mathrm{he}}\\ V_{\mathrm{MT}}=\mathrm{\&Sigma;}(P_e\left(t\right)\mathrm{\&Delta;t}/\left({\mathrm{\&eta;}}_e\left(t\right)L\right))\end{array}\right.$

In formula: Q _MT(t) be t combustion turbine exhaustion waste heat amount constantly; η _e(t) be the t generating efficiency of gas turbine constantly; η ₁Be gas turbine heat loss due to radiation coefficient; P _e(t) be the electrical power of t gas turbine output constantly; Q _he(t) heating capacity that provides for t moment gas turbine fume afterheat; K _heHeating efficiency for bromine cooling machine; V _MTAmount of natural gas for gas turbine consumption; Δ t is the running time of gas turbine; L is natural gas low-heat calorific value, gets 9.7kWh/m ³

The fuel cost calculating formula of miniature gas turbine is:

C _MT＝(C _n1/L)∑(P _e(t)Δt/η _e(t))

In formula: C _n1Be Gas Prices, this method is taken as 2.5 yuan/m ³

The example of the present embodiment is based on the C65 type miniature gas turbine of Capstone company, its η _e(t) and P _e(t) function referring to document " Study on economical operation of microgrid " (Li Le. Beijing: North China Electric Power University, 2011,3).

2, fuel cell mode

Depletion charge computing formula in the fuel cell power generation process is as follows:

C _FC＝(C _n1/L)∑(P _FC(t)Δt/η _FC(t))

In formula: P _FC(t), η _FC(t) be respectively power output and the gross efficiency of its t fuel cell constantly.

The example of the present embodiment is based on Proton Exchange Membrane Fuel Cells (40kW IFC PC-29), its η _FC(t) and P _FC(t) function is referring to document " Study on economical operation of microgrid ".

3, battery model

Storage battery be carved into during at constantly dump energy of t-1, t-1 at t dump energy constantly and storage battery t moment storage battery charge and discharge electric weight and electric weight attenuation hourly is relevant.

During battery discharging, P _SB(t) 〉=0, t residual capacity constantly is:

$\mathrm{SOC}\left(t\right)=\mathrm{SOC}(t-1)-P_{\mathrm{SB}}\left(t\right)/{\mathrm{\&eta;}}_D-D_B{Q}_B^S$

During charge in batteries, P _SB(t)≤0, t residual capacity constantly is:

$\mathrm{SOC}\left(t\right)=\mathrm{SOC}(t-1)-P_{\mathrm{SB}}\left(t\right){\mathrm{\&eta;}}_C-D_B{Q}_B^S$

In formula: SOC (t) is the t residual capacity of storage battery constantly; P _SB(t) for t constantly storage battery discharge and recharge power: η _C, η _DBe respectively charge and discharge efficient; D _BBe storage battery self discharge ratio hourly;

Total capacity for storage battery.

in the present embodiment, the model of exerting oneself of photovoltaic cell is referring to document " Study on economical operation when microgrid is incorporated into the power networks " (ox inscription, Huang Wei, Guo Jiahuan, Deng. electric power network technique, 2010,34 (11): 38-42), the model of exerting oneself of blower fan is referring to document " System modeling and optimization of microgrid using Genetic Algorithm " (Qijun Deng, Xing Gao, Hong Zhou.The2nd International Conference on Intelligent Control and Information Processing, 2011:540-544).

The multiple target economic dispatch model of this method comprises target function and constraints, and described objective function F (t) is intraday cost of electricity-generating F ₁(t) minimum, the Environmental costs F of pollutant to environmental impact simultaneously ₂(t) minimum, be specially:

minF(t)＝{F ₁(t)，F ₂(t)}t＝1，2...24 (1)

F ₁(t)＝C _f(t)+C _DP，(t)+C _OM(t)+C _grid(t)-C _sh(t) (2)

F ₂(t)＝C _c(t) (3)

In formula:

$C_f\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{f}_i\left(P_i\left(t\right)\right)---\left(4\right)$

$C_{\mathrm{DP}}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[\frac{C_{\mathrm{az},i}}{{8706k}_i}\&CenterDot;\frac{r{(1+r)}^{n_i}}{{(1+r)}^{n_i}-1}\&CenterDot;P_i\left(t\right)]---\left(5\right)$

$C_{\mathrm{OM}}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{K}_{\mathrm{OM},i}{\&CenterDot;P}_i\left(t\right)---\left(6\right)$

C _grid(t)＝CP(t)×PR _grid(t) (7)

C _sh(t)＝Q _he(t)×K _ph (8)

$C_e\left(t\right)=\underset{i=1}{\overset{n+1}{\mathrm{\&Sigma;}}}\left(\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}((V_{\mathrm{ej}}+V_j){\&CenterDot;Q}_{\mathrm{ij}}\&CenterDot;P_i\left(t\right))\right)---\left(9\right)$

C _f(t), C _DP(t), C _OM(t) be respectively t fuel cost, investment depreciable cost and the operation expense in each little source constantly; C _Grid(t) be that t moment microgrid is to the power purchase cost of outer net; C _sh(t) be the income that heats of t moment co-generation unit; C _e(t) be the constantly Environmental costs in each little source and the Environmental costs summation of the outside net purchase electricity of microgrid of t; f _iIt is the fuel cost function in i little source; P _i(t) be i little source t active power output constantly; N is the number in little source, and n+1 is interconnection power; C _{Az, i}It is the installation cost as per machine capacity in i little source; k _iBe the capacity factor in i little source, k _iAnnual energy output/(rated power in 8760 * this little source) in the=the i little source; R is Annual Percentage Rate; n _iIt is the pay back period of investment in i little source; K _{OM, i}It is the unit quantity of electricity operation expense coefficient in i little source; CP (t) is that t moment microgrid is to the purchase electricity price of outer net; PR _Grid(t) be that t moment microgrid is to the purchase of electricity of outer net; Q _he(t) be the t heating capacity of microgrid constantly; K _phPrice for the system of units heat; V _ejIt is the environmental value of j item pollutant; V _jIt is the suffered fine of j item pollutant; Q _ijIt is the j item pollutant discharge amount of i little source unit quantity of electricity; M is the kind of pollutant;

Described constraints comprises:

A, trend constraint:

${\mathrm{\&Delta;P}}_f=P_{\mathrm{fs}}-\underset{g\&Element;f}{\overset{h}{\mathrm{\&Sigma;}}}{V}_f{V}_g(G_{\mathrm{fg}}\cos {\mathrm{\&theta;}}_{\mathrm{fg}}+B_{\mathrm{fg}}\sin {\mathrm{\&theta;}}_{\mathrm{fg}})=0---\left(10\right)$

${\mathrm{\&Delta;Q}}_f=Q_{\mathrm{fs}}-\underset{g\&Element;f}{\overset{h}{\mathrm{\&Sigma;}}}{V}_f{V}_g(G_{\mathrm{fg}}\sin {\mathrm{\&theta;}}_{\mathrm{fg}}+B_{\mathrm{fg}}\cos {\mathrm{\&theta;}}_{\mathrm{fg}})=0---\left(11\right)$

In formula: h is the nodes of microgrid, f=1, and 2 ..., h; G _fg, B _fg, θ _fgBe respectively admittance and phase angle difference between node f and node g, g ∈ f represents the node that is connected with node f;

B, working voltage constraint:

V _f，min≤V _f≤V _f，max (12)

In formula; V _{F, min}, V _{F, max}Be respectively minimum, the maximum of the upper working voltage of node f;

The active power constraint of c, photovoltaic, blower fan and miniature gas turbine:

P _i，min≤P _i≤P _i，max (13)

In formula: P _{I, min}, P _{I, max}Be respectively minimum, the maximum of i little source active power;

D, the constraint of miniature gas turbine climbing rate:

P when increasing load _MT(t)-P _MT(t-1)≤R _{Up, MT}(14)

P during load shedding _MT(t-1)-P _MT(t)≤R _{Down, MT}(15)

In formula: P _MT(t) be the t active power output of miniature gas turbine constantly, R _{Up, MT}, R _{Down, MT}Be respectively the limit value of miniature gas turbine increase and reduction active power;

E, microgrid and outer net allow mutual through-put power constraint:

0≤P _Grid(t)≤P _Grid，max (16)

0≤Q _Grid(t)≤Q _Grid，max (17)

In formula: P _Grid(t), Q _Grid(t) be respectively t moment microgrid and outer net and allow mutual active power and reactive power; P _{Grid, max}, Q _{Grid, max}Be respectively and allow mutual maximum active power and reactive power;

Power factor for tie-line power transmission; C is the power factor limit value of tie-line power transmission;

Formula (16～18) shows the controllable burden of microgrid as external electrical network, guarantee that interconnection remains on certain scope to the power of microgrid transmission, and have higher power factor, guaranteed the fail safe of tie-line power transmission and more met electric company to dock requirement into microgrid;

F, storage battery operation constraint:

P _SB，min≤P _SB(t)≤P _SB，max (19)

-S _inv，SB≤P _SB(t)≤S _inv，SB (20)

$-\sqrt{S_{\mathrm{inv},\mathrm{SB}}^2-P_{\mathrm{SB}}^2\left(t\right)}\&le;Q_{\mathrm{SB}}\left(t\right)\&le;\sqrt{S_{\mathrm{inv},\mathrm{SB}}^2-P_{\mathrm{SB}}^2\left(t\right)}---\left(21\right)$

SOC _min≤SOC(t)≤SOC _max (22)

In formula: P _{SB, max}, P _{SB, min}Be respectively the minimum and maximum active power of storage battery; S _{Inv, SB}Capacity for the storage battery inverter; P _SB(t), Q _SB(t), SOC (t) be respectively t constantly the battery AC side discharge and recharge active power, reactive power and remaining battery capacity; SOC _min, SOC _maxBe respectively the minimum and maximum residual capacity of storage battery;

Formula (21) shows that the reactive power of storage battery output only is subjected to the restriction of its inverter capacity;

G, fuel cell operation constraint:

P _FC，min≤P _FC(t)≤P _FC，max (23)

P _FC(t)≤S _inv，FC (24)

$0\&le;Q_{\mathrm{FC}}\left(t\right)\&le;\sqrt{S_{\mathrm{inv},\mathrm{FC}}^2-P_{\mathrm{FC}}^2\left(t\right)}---\left(25\right)$

In formula: P _{FC, min}, P _{FC, max}Be respectively minimum, the maximum of fuel cell active power; P _FC(t), Q _FC(t) be respectively t active power and the reactive power of fuel cell AC constantly; S _{Inv, FC}Capacity for the fuel cell inverter;

Formula (20～21) and (24～25) show that the capacity priority allocation of storage battery and fuel cell inverter exerts oneself to meritorious;

H, spinning reserve constraint:

$P_{\mathrm{PV}}\left(t\right)+P_{\mathrm{WT}}\left(t\right)+P_{\mathrm{MT}}\left(t\right)+\min \{P_{\mathrm{SB},\max },\sqrt{{\left(S_{\mathrm{inv},\mathrm{SB}}\right)}^2-{\left(Q_{\mathrm{SB}}\left(t\right)\right)}^2},(\mathrm{SOC}(t-1)-\mathrm{SO}{C}_{\min }-$

(26)

$D_B{Q}_B^S){\mathrm{\&eta;}}_D\}+\min \{P_{\mathrm{FC},\max },\sqrt{{\left(S_{\mathrm{inv},\mathrm{FC}}\right)}^2-{\left(Q_{\mathrm{FC}}\left(t\right)\right)}^2}\}+P_{\mathrm{Grid},\max }>=P_D\left(t\right)+P_{\mathrm{loss}}\left(t\right)+R\left(t\right)$

In formula: P _PV(t), P _WT(t) be respectively the t active power of photovoltaic, blower fan constantly; P _D(t), P _Loss(t), R _D(t) be respectively t total load, total network loss and the required reserve capacity of micro-grid system constantly.

In multi-objective optimization question, multiobjective optimization solution and each sub-goal optimal solution are closely-related, should comprise the contribution of each sub-goal.But contradiction each other often between each sub-goal generally is difficult to make it to reach simultaneously optimum; Correlation between each sub-goal optimal solution and multiobjective optimization solution is all fuzzy, is difficult to a definite boundary, and can't directly compares its quality, therefore this method adopts fuzzy mathematics to solve multi-objective optimization question.

Find the solution thinking: first obtain the optimal solution of each sub-goal under institute's Prescribed Properties, recycle these optimal solutions with each sub-goal Function Modules gelatinization (namely determining membership function), then ask the membership function that makes common factor to get peaked solution, this solution is exactly the optimal solution of multi-objective optimization question.

Fuzzy mathematics is to show and process the blooming of actual objective reality with accurate mathematical method, reach this purpose, at first will determine membership function u.The size of u has reflected the satisfaction of optimum results, represents the most satisfiedly, and u=0 representative is least satisfied.

How to set up membership function, there is no so far unified approach can follow, and it is sought fado and depends on practical experience, finally to meet objective reality as standard.Membership function commonly used is divided into type less than normal and type bigger than normal, and for the former, target function is less, and satisfaction is larger, and the latter is just in time opposite.The multiple target of this method optimization is under the prerequisite of the Prescribed Properties satisfying, and reduces cost of electricity-generating and environmental contamination reduction, expects to have the upper limit and without lower limit, therefore selects the membership function of type less than normal.Type membership function less than normal commonly used has: falling half distributed rectangular, fall half trapezoidal profile, fall half Γ shape distribution, is 2 distributions and fall half distributed rectangular, is unsuitable for finding the solution of continuity optimization problem; Falling half trapezoidal profile needs first to determine the minimum and maximum value of target function, but target function peaked determine in the engineering of reality meaningless; Fall half Γ shape distribution therefore this method is selected, for described target function, membership function is:

$u_k\left(t\right)=\left\{\begin{array}{cc}1& F_k\left(t\right){\&le;F}_{k\min \left(t\right)}\\ \exp \left(\frac{F_{k\min }\left(t\right)-F_k\left(t\right)}{F_{k\min }\left(t\right)}\right)& F_k\left(t\right){>F}_{k\min \left(t\right)}\end{array}\right.---\left(27\right)$

In formula: F _Kmin(t) be single-goal function F _k(t) minimum value under described constraints, k=1,2; T=1,2 ..., 24.

After introducing the fuzzy membership variable, the non-linear single goal microgrid economic dispatch model after obfuscation is:

In formula: u (t), u ₁(t), u ₂(t) be respectively the satisfaction of fuzzy optimization, the satisfaction of cost of electricity-generating and the satisfaction of Environmental costs, wherein $u\left(t\right)=\underset{k=1}{\overset{2}{\mathrm{\&cap;}}}{u}_k\left(t\right).$

Genetic algorithm is a kind of random paralleling searching algorithm based on natural selection and gene genetics principle, is a kind of Finding Global Optimization and without any need for the effectively optimizing algorithm of initialization information.Crossover probability P in the parameter of genetic algorithm _cWith the variation probability P _mSelection be the key that affects performance of genetic algorithms, P _cLarger, the new individual speed that produces is faster, however P _cWhen excessive, the destroyed possibility of hereditary pattern is larger, P _cToo small, can make search procedure slow.For the variation probability P _m, P _mToo small, be not easy to produce new individual configurations, and P _mExcessive, genetic algorithm has become pure random search algorithm.M.Srinvivas etc. have proposed a kind of adaptive algorithm, P _cAnd P _mCan automatically change with fitness, but this algorithm approaches or when equaling maximum adaptation and spending P to the ideal adaptation degree _cAnd P _mApproach or equal zero, this is disadvantageous to the evolution initial stage, makes the defect individual at evolution initial stage be in a kind of state that does not change, and the possibility of moving towards locally optimal solution that causes evolving increases.

This method adopts Improving Genetic Algorithm: the crossover probability P in genetic algorithm _cWith the variation probability P _mCalculate according to following formula:

$P_c=\left\{\begin{array}{c}{P}_{c2}-\frac{(P_{c2}-P_{c3})(f^{\&prime;}-f_{\mathrm{avg}})}{f_{\max }-f_{\mathrm{avg}}}{f}^{\&prime;}{\&GreaterEqual;f}_{\mathrm{avg}}\\ P_{c1}-\frac{(P_{c1}-P_{c2})(f^{\&prime;}-f_{\min })}{f_{\mathrm{avg}}-f_{\min }}{f}^{\&prime;}{<f}_{\mathrm{avg}}\end{array}\right.---\left(34\right)$

$P_m=\left\{\begin{array}{c}{P}_{m2}-\frac{(P_{m2}-P_{m3})(f^{\&prime;}-f_{\mathrm{avg}})}{f_{\max }-f_{\mathrm{avg}}}{f}^{\&prime;}{\&GreaterEqual;f}_{\mathrm{avg}}\\ P_{m1}-\frac{(P_{m1}-P_{m2})(f^{\&prime;}-f_{\min })}{f_{\mathrm{avg}}-f_{\min }}{f}^{\&prime;}{<f}_{\mathrm{avg}}\end{array}\right.---\left(35\right)$

In formula: f _max, f _minAnd f _avgBe respectively maximum, minimum and average fitness in population; F ' is larger fitness in two individualities that will intersect, P _c1, P _c2, P _c3, P _m1, P _m2, P _m3Be the constant of setting, and P _c1, P _c2, P _c3∈ (0,1), P _m1, P _m2, P _m3∈ (0,0.1).

Improve the P that algorithm makes maximum adaptation degree individuality in colony _cAnd P _mNon-vanishing, bring up to respectively P _c3And P _m3, make defect individual not be in a kind of dead state, thereby make algorithm jump out locally optimal solution.And when individual fitness during less than average fitness, fitness value is more concentrated, makes algorithm easily be absorbed in locally optimal solution, and this improvement this moment has increased P _cAnd P _mThereby, increased population diversity.

The present embodiment further illustrates the application of this method take as shown in Figure 2 micro-grid system as example.The present embodiment uses and improves self-adapted genetic algorithm and taked roulette selection, multiple spot intersection and multiple spot to make a variation to find the solution following example.Population size is taken as 30, and iterations is 300.Because genetic algorithm belongs to the heuristic optimization algorithm, itself has randomness, and every suboptimization has different results, therefore the result that hereinafter provides is all the mean value that double counting obtains for 20 times.

In Fig. 2, A is resident load, maximum active power is 15kW, B is Commercial Load, and maximum active power is 30kW, and C is industrial load, maximum active power is 2 * 15kW, belong to interruptible load, D is industrial load, and maximum active power is 40kW, it is heat load that the power factor of three kinds of loads is all got 0.85, E.Consider the characteristics of the radial network configuration of distribution and low-voltage circuit parameter, line taking road resistance R=0.64 Ω/km, X=0.1 Ω/km.In Fig. 2,1～7 is node.

1.1 basic data

Heat income in the present embodiment and be taken as 0.1 yuan/kWh, the voltage permissible variation is-5%～+ 5%, meritorious and the idle upper limit of microgrid and outer net transmission is taken as respectively 50kW and 30.987kVar, the power factor limit value of tie-line power transmission is 0.85, the capacity of storage battery inverter is 60kVA, maximum, least residue capacity, initial capacity are respectively 100%, 30%, 70% rated capacity, and the rated capacity of storage battery is 900kWh, and the capacity of fuel cell inverter is 40kVA.The relevant information in each little source is as shown in table 1, WT, PV exert oneself, the daily load curve (with maximum electric loading percentage of this day) of heat load, the meritorious power curve (MT takes " electricity determining by heat " mode) of MT of being obtained by heat load and 3 kinds of character electric loadings as shown in Figure 3.Each little source pollutants emissions data, the pollutant standard of value, fine grade are referring to document, and Spot Price is referring to document.

The parameter in each little source of table 1

1.2 single goal optimum results

The present embodiment preferentially utilizes exerting oneself of WT and PV unit, follows the tracks of and controls maximum power output, supposes that MT does not provide idle exerting oneself.

1.2.1 cost of electricity-generating is minimum

When being incorporated into the power networks, because node 1 connects distribution, when calculating, trend regards it as balance node, get simultaneously its voltage as the reference voltage of microgrid, MT, FC and SB adopt power (PQ) to control, the meritorious optimum results take the cost of electricity-generating minimum as target as shown in Figure 4, the idle work optimization result is as shown in Figure 5, wherein meritorious vacancy=total burden with power-PV-WT-MT is meritorious exerts oneself, no-power vacancy=total load or burden without work.

In Fig. 4,1～7 period, in microgrid, load is lighter, and the dump energy that PV, WT and MT send charges a battery, 8～17 periods, there is meritorious vacancy in system, preferentially calls the meritorious of storage battery and exerts oneself, therefore storage battery almost is in the full capacity generating state, 17 period Mo storage battery residual capacity reached lower limit, still there is meritorious vacancy in 18～20 period microgrids, and the meritorious dump energy that is 0,21～24 period PV, WT and MT send of storage battery output charges a battery.1～9 and 21～24 periods, storage battery just can satisfy the microgrid safe and reliable operation (on the basis of cutting load not in the scope of exerting oneself, microgrid can move under Prescribed Properties satisfying), therefore outwards net purchase is not electric, 10～13 and 15～20 periods, can not satisfy the microgrid safe and reliable operation in the scope of exerting oneself of storage battery, and this moment purchase electricity price lower than the cost of electricity-generating of FC, therefore preferentially satisfy the microgrid power shortage from the outer net power purchase.14 period PV, WT, MT and SB can't satisfy the microgrid safe and reliable operation, and this moment purchase electricity price higher than the cost of electricity-generating of FC, preferentially call FC meritorious vacancy is provided, the outside net purchase electricity of insufficient section satisfies, other periods need not called FC provides meritorious exerting oneself just can satisfy the microgrid safe and reliable operation, provides idle and exerts oneself therefore only call FC.

As can be seen from Figure 4, before 17 periods, the meritorious vacancy of microgrid is mainly provided by SB, meritorious vacancy is mainly provided by outer net after 17 period Mo SB dump energies arrive lower limit, tie-line power transmission has mainly played the effect of peak regulation here, and FC only just provides meritorious in cost of electricity-generating during lower than purchase electricity price and exerts oneself.

As can be seen from Figure 5, it is idle that SB and FC provide on preferential satisfied meritorious Demand Base, coordinates the idle reactive requirement that satisfies microgrid of interconnection.The reactive requirement of micro-grid system is mainly provided by SB and FC, microgrid is only to need outer net to provide the meritorious period (10～20 period) of exerting oneself to satisfy the microgrid safe and reliable operation just to absorb from outer net a small amount of idle, and outer net is higher to the power factor of microgrid through-put power, as a controllable burden access distribution, distribution requires to conform to the power factor of access load with microgrid.

1.2.2 Environmental costs are minimum

Take the minimum meritorious optimum results as target of Environmental costs as shown in Figure 6, the idle work optimization result as shown in Figure 7.

In Fig. 6,1～7 period, in microgrid, load is lighter, the dump energy that PV, WT and MT send charges a battery, and 8～20 periods, there is meritorious vacancy in system, preferentially call the meritorious of storage battery and exert oneself, the dump energy that 21～24 period PV, WT and MT send charges a battery.1～9 and 17～24 periods, storage battery can satisfy the microgrid safe and reliable operation in the scope of exerting oneself, therefore outside net purchase electricity or call the meritorious of FC and exert oneself, 10～16 periods, can't satisfy the microgrid safe and reliable operation in the scope of exerting oneself of storage battery, the preferential lower-cost FC of transfer environment provides meritorious and exerts oneself, therefore FC almost is in the full capacity generating state, the outside net purchase electricity of insufficient section satisfies.

As can be seen from Figure 6, the meritorious vacancy of microgrid is mainly provided by SB, and FC and interconnection power have mainly played the effect of peak regulation here, and environmental benefit preferably FC born main peak regulation task.

As can be seen from Figure 7, it is idle that SB and FC provide on preferential satisfied meritorious Demand Base, coordinates the idle reactive requirement that satisfies microgrid of interconnection transmission.The reactive requirement of micro-grid system is mainly provided by SB and FC, because the Environmental costs of outside net purchase electricity are higher, therefore microgrid is only needing outer net to provide the meritorious period (10～16 period) of exerting oneself to satisfy the microgrid safe and reliable operation just to absorb the idle of minute quantity from outer net.

1.3 multiple-objection optimization result

The optimum results 1.3.1 exert oneself

With the meritorious optimum results after multi objective fuzzy as shown in Figure 8, the idle work optimization result as shown in Figure 9.

Under multiple target, the charging and discharging state of day part storage battery is similar to single goal as can be seen from Figure 8.1～7 period, in microgrid, load is lighter, and the dump energy that PV, WT and MT send charges a battery, 8～20 periods, there is meritorious vacancy in system, preferentially calls the meritorious of storage battery and exerts oneself, and the dump energy that 21～24 period PV, WT and MT send charges a battery.1～9 and 17～24 periods, storage battery can satisfy the microgrid safe and reliable operation in the scope of exerting oneself, therefore outside net purchase electricity or call the meritorious of FC and exert oneself, 10～16 periods, can't satisfy the microgrid safe and reliable operation in the scope of exerting oneself of storage battery, consider that cost of electricity-generating and Environmental costs are called FC and interconnection power satisfies the microgrid safe and reliable operation.

As can be seen from Figure 8, the meritorious vacancy of microgrid is mainly provided by SB, FC and interconnection power shared the task of peak regulation.

As can be seen from Figure 9, it is idle that SB and FC provide on preferential satisfied meritorious Demand Base, coordinates the idle reactive requirement that satisfies microgrid of interconnection transmission.The reactive requirement of micro-grid system is mainly provided by SB and FC, and the idle of SB and FC became complementary fit to satisfy the reactive requirement of microgrid in one day substantially.Microgrid is only to need outer net to provide the meritorious period (10～16 period) of exerting oneself to satisfy the microgrid safe and reliable operation just to absorb from outer net a small amount of idle.

1.3.2 single goal and multiple-target system optimal value are relatively

Every total cost that single goal and multiple-objection optimization draw is as shown in table 2.

Expense under table 2 different target

As can be seen from Table 2, the Environmental costs that draw take the cost of electricity-generating minimum as target are higher by 15.6% than minimum Environmental costs, and environmental benefit is relatively poor; And higher by 6.2% than minimum cost of electricity-generating take the minimum cost of electricity-generating that draws as target of Environmental costs, economy is relatively poor; The cost of electricity-generating that multi-target fuzzy optimal draws is higher by 3.9% than minimum cost of electricity-generating, and Environmental costs are higher by 1.1% than minimum Environmental costs, has synthesized and coordinated economy and the feature of environmental protection.

The present embodiment converts the pollutant unification that produces in the power generation process of little source to environmental benefit that Environmental costs are weighed the microgrid operation, has represented the integrated value of pollutant discharge amount.

Single goal and multiple-objection optimization draw each constantly cost of electricity-generating as shown in figure 10, Environmental costs are as shown in figure 11.

Can find out from Figure 10 and Figure 11, in multi-target fuzzy optimal draw a day, cost of electricity-generating and Environmental costs are a little more than minimum cost of electricity-generating and minimum Environmental costs.And all higher than minimum cost of electricity-generating and minimum Environmental costs take the minimum cost of electricity-generating that draws as target of Environmental costs and the Environmental costs that draw take the cost of electricity-generating minimum as target.Multi-target fuzzy optimal has been coordinated economy and environment balance between the two preferably, microgrid owner and user's interests and social benefit have been taken into account, not only improved power supply quality, and taken into account the utilization of environmental protection and regenerative resource, the economic dispatch that has realized distributed power source with rationally utilize.In sum, multi-objective Model more can reflect the practical operation situation of microgrid definitely than single goal model, and this model can reach environmental protection effect preferably with the little Financial cost of trying one's best, and has realized the energy-conserving and environment-protective power generation dispatching.

1.3.3 satisfaction optimum results

After multi objective fuzzy, the satisfaction of each sub-goal and the satisfaction of multi objective fuzzy are as shown in figure 12.

As can be seen from Figure 12, each sub-goal satisfaction and the fuzzy optimization satisfaction of multiple-objection optimization are higher on the whole.Can find out from the satisfaction curve of each sub-goal, what the satisfaction of multi-target fuzzy optimal was had appreciable impact is the satisfaction of cost of electricity-generating, is secondly the satisfaction of Environmental costs.This means affects in all of microgrid power generation dispatching factors, and economy is decisive factor, is secondly environmental factor.This is because the power supply in microgrid mostly is greatly clean energy resource, although the little source emission pollutant that has with respect to conventional power source, is produced the pollutant that every kilowatt hour electric weight discharges lower.

energy control method of the present invention has been set up and has been considered that little source provides simultaneously meritorious and idle and exerts oneself and take into account the cogeneration of heat and power type micro-grid system multiple target economic dispatch model that heats income, comprise wind with one, light, storage, miniature gas turbine, the concrete microgrid of fuel cell and thermoelectricity load is example, optimized the meritorious of each little source under the mode that is incorporated into the power networks of considering Spot Price by fuzzy theory and improved genetic algorithms method, idle exerting oneself and the satisfaction of multiple-objection optimization, comparative analysis single goal and multiple-target system optimal value, simulation example shows the actual motion state that multi-objective Model more can the precise reaction microgrid than single goal model, can realize environmental benefit preferably with the little operating cost of trying one's best, and the unit (as storage battery) of good combination property has obvious advantage when considering the multiple targets such as economy and the feature of environmental protection, verified the validity of the model and algorithm of putting forward.

Control the meritorious and idle output in each little source in microgrid by control method of the present invention, can make little source itself provide idle exerting oneself to coordinate outer net idlely to exert oneself to satisfy the System Reactive Power demand to what microgrid provided, the additional investment of having fallen less reactive power compensator is installed, guarantee that outer net need not excessive reactive power support to be provided and to guarantee that the power that transmits between outer net and microgrid has higher power factor to microgrid, meet better electric company and dock requirement into microgrid.

Claims (6)

1. one kind based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, described microgrid comprises a plurality of little sources and load, each little source connects by interconnection with being connected, described load comprises electric loading and heat load, it is characterized in that, described method specifically comprises the following steps:

1) processor obtains the information in each little source and interconnection in microgrid, sets up multiple target economic dispatch model take cost of electricity-generating and Environmental costs minimum as target;

2) set up membership function, multiple target economic dispatch model is carried out obfuscation, obtain non-linear single goal microgrid economic dispatch model;

3) adopt Improving Genetic Algorithm that non-linear single goal microgrid economic dispatch model is optimized and find the solution, obtain the meritorious and reactive power data of the optimum of each little source and interconnection, and with this transfer of data to the microgrid power controller;

4) the microgrid power controller is gained merit and the meritorious of each little source of reactive power Data Control and interconnection and reactive power output according to the optimum that receives.

2. according to claim 1 a kind ofly it is characterized in that based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, described little source comprises photovoltaic, blower fan, miniature gas turbine, fuel cell and storage battery.

3. according to claim 2 a kind of based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, it is characterized in that, described multiple target economic dispatch model comprises target function and constraints, and described objective function F (t) is intraday cost of electricity-generating F ₁(t) minimum, the Environmental costs F of pollutant to environmental impact simultaneously ₂(t) minimum, be specially:

min F(t)＝{F ₁(t)F ₂(t)}t＝1，2...24 (1)

F ₁(t)＝C _f(t)+C _DP(t)+C _OM(t)+C _grid(t)-C _sh(t) (2)

F ₂(t)＝C _e(t) (3)

In formula:

$C_f\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{f}_i\left(P_i\left(t\right)\right)---\left(4\right)$

$C_{\mathrm{DP}}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[\frac{C_{\mathrm{az},i}}{{8706k}_i}\&CenterDot;\frac{r{(1+r)}^{n_i}}{{(1+r)}^{n_i}-1}\&CenterDot;P_i\left(t\right)]---\left(5\right)$

$C_{\mathrm{OM}}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{K}_{\mathrm{OM},i}{\&CenterDot;P}_i\left(t\right)---\left(6\right)$

C _grid(t)＝CP(t)×PR _grid(t) (7)

C _sh(t)＝Q _he(t)×K _ph (8)

$C_e\left(t\right)=\underset{i=1}{\overset{n+1}{\mathrm{\&Sigma;}}}\left(\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}((V_{\mathrm{ej}}+V_j){\&CenterDot;Q}_{\mathrm{ij}}\&CenterDot;P_i\left(t\right))\right)---\left(9\right)$

C _f(t), C _DP(t), C _OM(t) be respectively t fuel cost, investment depreciable cost and the operation expense in each little source constantly; C _Grid(t) be that t moment microgrid is to the power purchase cost of outer net; C _sh(t) be the income that heats of t moment co-generation unit; C _e(t) be t Environmental costs and the Environmental costs summation of the outside net purchase electricity of microgrid: the f in each little source constantly _iIt is the fuel cost function in i little source; P _i(t) be i little source t active power output constantly; N is the number in little source, and n+1 is interconnection power; C _{Az, i}It is the installation cost as per machine capacity in i little source; k _iBe the capacity factor in i little source, k _iAnnual energy output/(rated power in 8760 * this little source) in the=the i little source; R is Annual Percentage Rate; n _iIt is the pay back period of investment in i little source; K _{OM, i}It is the unit quantity of electricity operation expense coefficient in i little source; CP (t) is that t moment microgrid is to the purchase electricity price of outer net; PR _Grid(t) be that t moment microgrid is to the purchase of electricity of outer net; Q _he(t) be the t heating capacity of microgrid constantly; K _phPrice for the system of units heat; V _ejIt is the environmental value of j item pollutant; V _jIt is the suffered fine of j item pollutant; Q _ijIt is the j item pollutant discharge amount of i little source unit quantity of electricity; M is the kind of pollutant;

Described constraints comprises:

A, trend constraint:

${\mathrm{\&Delta;P}}_f=P_{\mathrm{fs}}-\underset{g\&Element;f}{\overset{h}{\mathrm{\&Sigma;}}}{V}_f{V}_g(G_{\mathrm{fg}}\cos {\mathrm{\&theta;}}_{\mathrm{fg}}+B_{\mathrm{fg}}\sin {\mathrm{\&theta;}}_{\mathrm{fg}})=0---\left(10\right)$

${\mathrm{\&Delta;Q}}_f=Q_{\mathrm{fs}}-\underset{g\&Element;f}{\overset{h}{\mathrm{\&Sigma;}}}{V}_f{V}_g(G_{\mathrm{fg}}\sin {\mathrm{\&theta;}}_{\mathrm{fg}}+B_{\mathrm{fg}}\cos {\mathrm{\&theta;}}_{\mathrm{fg}})=0---\left(11\right)$

In formula: h is the nodes of microgrid, f=1, and 2 ..., h; G _fg, B _fg, θ _fgBe respectively admittance and phase angle difference between node f and node g, g ∈ f represents the node that is connected with node f;

B, working voltage constraint:

V _f，min≤V _f≤V _f，max (12)

In formula: V _{F, min}, V _{F, max}Be respectively minimum, the maximum of the upper working voltage of node f;

The active power constraint of c, photovoltaic, blower fan and miniature gas turbine:

P _i，min≤P _i≤P _i，max (13)

In formula: P _{I, min}, P _{I, max}Be respectively minimum, the maximum of i little source active power;

D, the constraint of miniature gas turbine climbing rate:

P when increasing load _MT(t)-P _MT(t-1)≤R _{Up, MT}(14)

P during load shedding _MT(t-1)-P _MT(t)≤R _{Dowm, MT}(15)

In formula: P _MT(t) be the t active power output of miniature gas turbine constantly, R _{Up, MT}, R _{Down, MT}Be respectively the limit value of miniature gas turbine increase and reduction active power;

E, microgrid and outer net allow mutual through-put power constraint:

0≤P _Grid(t)≤P _Grid，max (16)

0≤Q _Grid(t)≤Q _Grid，max (17)

In formula: P _Grid(t), Q _Grid(t) be respectively t moment microgrid and outer net and allow mutual active power and reactive power; P _{Grid, max}, Q _{Grid, max}Be respectively and allow mutual maximum active power and reactive power;

Power factor for tie-line power transmission; C is the power factor limit value of tie-line power transmission;

F, storage battery operation constraint:

P _SB，min≤P _SB(t)≤P _SB，max (19)

-S _inv，SB≤P _SB(t)≤S _inv，SB (20)

$-\sqrt{S_{\mathrm{inv},\mathrm{SB}}^2-P_{\mathrm{SB}}^2\left(t\right)}\&le;Q_{\mathrm{SB}}\left(t\right)\&le;\sqrt{S_{\mathrm{inv},\mathrm{SB}}^2-P_{\mathrm{SB}}^2\left(t\right)}---\left(21\right)$

SOC _min≤SOC(t)≤SOC _max (22)

In formula: P _{SB, max}, P _{SB, min}Be respectively the minimum and maximum active power of storage battery; S _{Inv, SB}Capacity for the storage battery inverter; P _SB(t), Q _SB(t), SOC (t) be respectively t constantly the battery AC side discharge and recharge active power, reactive power and remaining battery capacity; SOC _min, SOC _maxBe respectively the minimum and maximum residual capacity of storage battery;

G, fuel cell operation constraint:

P _FC，min≤P _FC(t)≤P _FC，max (23)

P _FC(t)≤S _inv，FC (24)

$0\&le;Q_{\mathrm{FC}}\left(t\right)\&le;\sqrt{S_{\mathrm{inv},\mathrm{FC}}^2-P_{\mathrm{FC}}^2\left(t\right)}---\left(25\right)$

In formula: P _{FC, min}, P _{FC, max}Be respectively minimum, the maximum of fuel cell active power; P _FC(t), Q _FC(t) be respectively t active power and the reactive power of fuel cell AC constantly; S _{Inv, FC}Capacity for the fuel cell inverter;

H, spinning reserve constraint:

$P_{\mathrm{PV}}\left(t\right)+P_{\mathrm{WT}}\left(t\right)+P_{\mathrm{MT}}\left(t\right)+\min \{P_{\mathrm{SB},\max },\sqrt{{\left(S_{\mathrm{inv},\mathrm{SB}}\right)}^2-{\left(Q_{\mathrm{SB}}\left(t\right)\right)}^2},(\mathrm{SOC}(t-1)-\mathrm{SO}{C}_{\min }-$

(26)

$D_B{Q}_B^S){\mathrm{\&eta;}}_D\}+\min \{P_{\mathrm{FC},\max },\sqrt{{\left(S_{\mathrm{inv},\mathrm{FC}}\right)}^2-{\left(Q_{\mathrm{FC}}\left(t\right)\right)}^2}\}+P_{\mathrm{Grid},\max }>=P_D\left(t\right)+P_{\mathrm{loss}}\left(t\right)+R\left(t\right)$

In formula: P _PV(t), P _WT(t) be respectively the t active power of photovoltaic, blower fan constantly; η _DBe discharging efficiency; D _BBe storage battery self discharge ratio hourly;

Total capacity for storage battery; P _D(t), P _Loss(t), R _D(t) be respectively t total load, total network loss and the required reserve capacity of micro-grid system constantly.

4. according to claim 3 a kind of based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, it is characterized in that, described membership function adopts and falls the membership function that half Γ shape distributes, and for described target function, membership function is:

$u_k\left(t\right)=\left\{\begin{array}{cc}1& F_k\left(t\right){\&le;F}_{k\min \left(t\right)}\\ \exp \left(\frac{F_{k\min }\left(t\right)-F_k\left(t\right)}{F_{k\min }\left(t\right)}\right)& F_k\left(t\right){>F}_{k\min \left(t\right)}\end{array}\right.---\left(27\right)$

In formula: F _Kmin(t) be single-goal function F _k(t) minimum value under described constraints, k=1,2; T=1,2 ..., 24.

5. according to claim 4 a kind ofly it is characterized in that based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, the non-linear single goal microgrid economic dispatch model after described obfuscation is:

In formula: u (t), u ₁(t), u ₂(t) be respectively the satisfaction of fuzzy optimization, the satisfaction of cost of electricity-generating and the satisfaction of Environmental costs, wherein $u\left(t\right)=\underset{k=1}{\overset{2}{\mathrm{\&cap;}}}{u}_k\left(t\right).$

6. according to claim 5 a kind ofly it is characterized in that based on microgrid energy control method under the mode that is incorporated into the power networks of Multiobjective Scheduling model, described Improving Genetic Algorithm refers to: the crossover probability P in genetic algorithm _cWith the variation probability P _mCalculate according to following formula:

$P_c=\left\{\begin{array}{c}{P}_{c2}-\frac{(P_{c2}-P_{c3})(f^{\&prime;}-f_{\mathrm{avg}})}{f_{\max }-f_{\mathrm{avg}}}{f}^{\&prime;}{\&GreaterEqual;f}_{\mathrm{avg}}\\ P_{c1}-\frac{(P_{c1}-P_{c2})(f^{\&prime;}-f_{\min })}{f_{\mathrm{avg}}-f_{\min }}{f}^{\&prime;}{<f}_{\mathrm{avg}}\end{array}\right.---\left(29\right)$

$P_m=\left\{\begin{array}{c}{P}_{m2}-\frac{(P_{m2}-P_{m3})(f^{\&prime;}-f_{\mathrm{avg}})}{f_{\max }-f_{\mathrm{avg}}}{f}^{\&prime;}{\&GreaterEqual;f}_{\mathrm{avg}}\\ P_{m1}-\frac{(P_{m1}-P_{m2})(f^{\&prime;}-f_{\min })}{f_{\mathrm{avg}}-f_{\min }}{f}^{\&prime;}{<f}_{\mathrm{avg}}\end{array}\right.---\left(30\right)$

In formula: f _max, f _minAnd f _avgBe respectively maximum, minimum and average fitness in population; F ' is larger fitness in two individualities that will intersect, P _c1, P _c2, P _c3, P _m1, P _m2, P _m3Be the constant of setting, and P _c1, P _c2, P _c3∈ (0,1), P _m1, P _m2, P _m3∈ (0,0.1).

Images