CN104362681A - Island micro-grid capacity optimal-configuration method considering randomness - Google Patents

Abstract

The invention relates to the field of micro-grids, in particular to an island micro-grid capacity optimal-configuration method considering randomness. According to the island micro-grid capacity optimal-configuration method considering the randomness, a probability model of random meteorological factors is analyzed, a probability density distribution function is obtained, distribution parameters are obtained through Monte Carlo simulation, and finally active power output and electrical load of various distributed power supplies are obtained. The cost minimization serves as an objective function, the unit installing number limitation, micro-grid internal electric power and energy balance, comprehensive coordination reliability and environmental protection factors serve as constraint conditions, and solution is performed by means of genetic algorithm to obtain the number of various distributed power supplies in an optimal allocation scheme and the operation time of a diesel generator. In the island micro-grid capacity optimal-configuration method, the power output randomness of the distributed power supplies in micro-grids and the economical efficiency, the reliability, the environmental protection factors and other factors of the micro-grids are considered, a micro-grid allocation scheme which is relatively low in cost and relatively high in reliability is established, and on the premise that electrical load is met, the cost is saved, and unnecessary energy waste is avoided.

Description

A kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness

Technical field

The present invention relates to micro-capacitance sensor field, particularly relate to a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness.

Background technology

There is the electrical network of a large amount of annual islet operation in remote districts, its internal electric source is often made up of single diesel engine generator.Along with the day by day in short supply of fossil fuel and the reinforcement gradually of Public environmental attitude, be necessary to build renewable energy power generation in the electrical network of islet operation.Meanwhile, along with the exploitation of remote districts, power supply reliability may be there is in islet operation micro-capacitance sensor and require higher user, therefore be necessary to improve wherein power supply architecture, use rational power configuration scheme, meet the diversified reliability requirement of customization.

Micro-capacitance sensor capacity optimization allocation is one of key issue of micro-capacitance sensor program and design, traditional analytical method is that inner for micro-capacitance sensor distributed power source is equivalent to traditional power supply node type, even being exerted oneself by distributed power source, it is constant to be set to, such solution problem is have ignored the randomness that in micro-capacitance sensor, distributed power source is exerted oneself, and thus cannot consider micro-capacitance sensor self-characteristic when micro-capacitance sensor capacity is distributed rationally.In addition, micro-capacitance sensor capacity optimization allocation relates to the multifactorial coordinations such as economy, reliability and environmental protection, therefore, sets up and can consider that the mathematic optimal model of these factors is particularly important.

Summary of the invention

The object of the present invention is to provide a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness.It can consider that randomness that in micro-capacitance sensor, distributed power source is exerted oneself, the economy of micro-capacitance sensor, reliability and environmental protection etc. are multifactor, set up a set of advantage of lower cost, the micro-capacitance sensor allocation plan that reliability is relatively high, under the prerequisite meeting power load, save cost, avoid unnecessary energy waste, and carbon emission amount is lower, little to environmental impact.

Technical scheme of the present invention is: a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness, comprises the following steps:

In a, analysis micro-capacitance sensor, the relation of all kinds of distributed power source and randomness meteorologic factor, sets up the probabilistic model of the randomness meteorologic factor relevant to all kinds of distributed power source and draws its probability density function;

B, meritoriously exerting oneself and the relational expression of randomness meteorologic factor of calculating all kinds of distributed power source, this relational expression is brought in the probability density function of the randomness meteorologic factor relevant with all kinds of distributed power source, show that all kinds of distributed power source is gained merit the probability density function of exerting oneself;

C, historical statistical data according to the randomness meteorologic factor relevant with all kinds of distributed power source, show that all kinds of distributed power source is gained merit the distributed constant of the probability density function of exerting oneself;

D, to gain merit according to all kinds of the distributed power source probability density function and its distributed constant of exerting oneself, random the meritorious of all kinds of distributed power source that produce is exerted oneself;

E, draw the generated output of micro-capacitance sensor according to meritorious the exerting oneself of all kinds of distributed power source;

F, obtain the probability density function of power load by normal distribution;

G, historical statistical data according to power load, draw the distributed constant of its probability density function;

H, produce power load at random according to the probability density function of power load and the distributed constant of probability density function;

I, based on all kinds of distributed power source, set up cost minimum target function, and using the quantity of the generated output of micro-capacitance sensor and power load Real-time Balancing, all kinds of distributed power source, power supply reliability and Environmental Factors as constraints;

J, with cost minimization target function for fitness function, with the installation number of all kinds of distributed power source for decision variable, the installation number of all kinds of distributed power source when going out cost minimization by genetic algorithm for solving under the screening restriction of constraints.

K, carry out the capacity configuration of whole micro-capacitance sensor according to the installation number of all kinds of distributed power source.

Further, described all kinds of distributed power source comprises diesel engine generator, wind-driven generator and/or photovoltaic battery matrix.

Further, the randomness meteorologic factor that described and all kinds of distributed power source is relevant comprises wind speed and intensity of illumination.

Further, the gain merit distributed constant of the probability density function of exerting oneself, the distributed constant of described power load power probability density distribution function of described all kinds of distributed power source is all obtained by Monte Carlo simulation.

Further, in step e, the generated output of micro-capacitance sensor is N _wG.P _wG+ N _pV.P _pV+ N _dG.P _dG, wherein, N _wG, N _pV, N _dGbe respectively the quantity of wind-driven generator, photovoltaic battery matrix, diesel engine generator, P _wG, P _pV, P _dGbe respectively wind-driven generator, photovoltaic battery matrix, the meritorious of diesel engine generator exert oneself.

Further, cost minimization target function C in step I _Σ=(N _wG.P _wG.C _wG+ N _pV.P _pV.C _pV+ N _dG.P _dG.C _dG) _min, C in formula _wG, C _pV, C _dGbe respectively the cost of each wind-driven generator, photovoltaic battery matrix, diesel engine generator.

Further, in step I, the generated output of micro-capacitance sensor and the Real-time Balancing expression formula of power load power are: $N_{\mathrm{WG}}\·P_{\mathrm{WG}.t}+N_{\mathrm{PV}}\·P_{\mathrm{PV}.t}+N_{\mathrm{DG}}\·P_{{\mathrm{DG}}^{.t}}=P_{L.t},$ In formula, P _{wG, t}, P _{pV, t}, P _{dG, t}be respectively meritorious when time t of wind-driven generator, photovoltaic battery matrix, diesel engine generator to exert oneself, P _l,tfor power load during t.

Further, in step I, the quantity of all kinds of distributed power source is selected according to the scale in place, and its span is N _i ^min<=N _i<=N _i ^max, with be respectively all kinds of distributed power source arranging in place the minimum number and maximum quantity that can arrange.

Further, in step I, the judge index of power supply reliability is minimum short of electricity probability LOLP, and minimum short of electricity probability LOLP is not more than the maximum short of electricity probable value LOLP of artificial setting ^max.

Further, in step I, Environmental Factors is the discharge capacity of CO2, and the computing formula of CO2 discharge capacity is: CO ₂(P _dG)=a+bP _dG+ cP _dG ², in formula, a, b, c are empirical coefficient, by setting the power P that acceptable maximum CO2 discharge capacity is come diesel engine generator _dGlimit.

Further, the cost C of described diesel engine generator _dGcomprise unit cost, installation cost and operating cost, described operating cost Cd (P _dG)=k1+k2P _dG+ k3P _dG ², wherein, k1, k2, k3 are empirical coefficient, and k1 span is 1.0165 ~ 1.1235, k2 span be 0.0624 ~ 0.0690, k3 span is 0.000057 ~ 0.000063.

The invention has the beneficial effects as follows: method proposed by the invention can consider the randomness that in micro-capacitance sensor, distributed power source is exerted oneself, and the economy of comprehensive micro-capacitance sensor, reliability and environmental protection etc. are multifactor, solve a set of preferably micro-capacitance sensor allocation plan.Good economic benefit, advantage of lower cost and there is higher reliability can be had under the prerequisite meeting power load according to micro-capacitance sensor that this cover allocation plan is arranged.Adopt computational methods of the present invention, the electric energy that micro-capacitance sensor provides is meeting the quantity of layout diesel engine generator the least possible on the basis of minimum short of electricity probability LOLP, is reducing the running time of diesel engine generator as far as possible, and maximized utilization uses the wind-driven generator of renewable energy resources driving and photovoltaic battery matrix to provide electric energy.Meanwhile, also according to the ratio of the natural conditions reasonable distribution wind-driven generator of micro-capacitance sensor affiliated area and photovoltaic battery matrix, electric energy can be exported and maximizes.So; then whole micro-capacitance sensor can when the constraints of satisfied setting; avoid unnecessary energy waste; and owing to decreasing the running time of diesel engine generator as much as possible; while the use decreasing conventional fossil fuel, decrease the discharge capacity of greenhouse gas CO2, serve the effect of certain protection of the environment; more save cost owing to saving the quantity of fuel oil and diesel engine, there is good economic benefit.Meanwhile, add more clean wind power generation and photovoltaic solar generating in micro-capacitance sensor after, the selectivity of user to electric power supply increases, and is conducive to promoting power supply level, improves user power utilization satisfaction.

Accompanying drawing explanation

Fig. 1 is monthly average wind speed and intensity of illumination situation;

Fig. 2 is monthly average power consumption situation;

Fig. 3 is year wind speed simulation situation;

Fig. 4 is year illumination strength simulation situation;

Fig. 5 is year power load analog case

Fig. 6 is iterations and cost relation in genetic algorithm;

Fig. 7 is reliability index and Cost sensitivity relation;

Fig. 8 is the capacity configuration optimizing method realization flow figure of micro-capacitance sensor;

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Specific embodiment described herein only in order to explain the present invention, is not intended to limit the present invention.In addition, if below in described each execution mode of the present invention involved technical characteristic do not form conflict each other and just can mutually combine.

In this computational methods of the present invention, all kinds of distributed power source has multiple, can be the generating set that diesel engine generator and any regenerative resource drive, for convenience of description, be described in the present embodiment with wind-driven generator, photovoltaic battery matrix and diesel engine generator.

Wind-driven generator relies on wind-force to generate electricity, so the factor affecting its generating randomness is wind speed.Be illustrated in figure 1 average wind conditions monthly in a year.According to general knowledge, we know that Weibull distribution is generally considered and surveys the good probabilistic model of wind speed profile matching, and according to this probabilistic model, we can show that the probability density function of wind speed is:

$f\left(v\right)=\frac{\mathrm{\&alpha;}}{\mathrm{\&beta;}}{\left(\frac{v}{\mathrm{\&beta;}}\right)}^{\mathrm{\&alpha;}-1}\exp [-{\left(\frac{v}{p}\right)}^{\mathrm{\&alpha;}}]---\left(1\right)$

In formula: v is wind speed; α > 0 is form parameter, and β > 0 is scale parameter; α and β can obtain (mean value and standard deviation data are as accompanying drawing 1) according to the average value mu of wind speed and standard deviation δ, and the corresponding formula of α and β is:

$\mathrm{\&alpha;}={\left(\frac{\mathrm{\&delta;}}{\mathrm{\&mu;}}\right)}^{-1.086}---\left(2\right)$

$\mathrm{\&beta;}=\frac{\mathrm{\&mu;}}{\mathrm{\&Gamma;}(1+\frac{1}{\mathrm{\&alpha;}})}---\left(2\right)$

Wherein, Γ represents gamma function.

When known wind speed, utilize wind-driven generator to gain merit to exert oneself the relation with wind speed, can show that wind-driven generator is gained merit the formula of exerting oneself:

P in formula _rfor the rated power of wind turbine generator; v _rfor rated wind speed; v _cifor incision wind speed; v _cofor cut-out wind speed.

The relational expression of exerting oneself of being gained merit by wind-driven generator in formula (4) is brought in the probability-distribution function of wind speed, and by can obtain the meritorious probability-distribution function of exerting oneself of wind-driven generator in each piecewise function upper integral, its formula is as follows:

As shown in Figure 3, the average value mu of wind speed and standard deviation δ are obtained by Monte Carlo simulation.The method utilizes local longitude and latitude to inquire about monthly mean wind speed at NSDR weather site, generate wind speed per hour, thus average value mu and the standard deviation δ of wind speed every day can be obtained, the value of α and β both can have been drawn according to formula (2) and formula (3).Recycling formula (1), air speed data is produced at random according to Monte Carlo method setting probability, and calculate the random meritorious of wind-driven generator by formula (4) and exert oneself, take in its probability density function formula (5), both can produce wind-driven generator at random according to setting probability and to gain merit the data of exerting oneself.

Same method is also used when calculating photovoltaic battery matrix and gaining merit and exert oneself.The factor affecting photovoltaic battery matrix generating randomness is intensity of illumination, and through statistics, in known 1 year, the distribution situation of intensity of illumination as shown in Figure 1.According to statistics, in a few hours yardsticks, intensity of illumination can be similar to regards beta distribution as, and its probability density function is as follows:

$f\left(r\right)=\frac{\mathrm{\&Gamma;}({\mathrm{\&alpha;}}^{\&prime;}+{\mathrm{\&beta;}}^{\&prime;})}{\mathrm{\&Gamma;}\left({\mathrm{\&alpha;}}^{\&prime;}\right)\mathrm{\&Gamma;}\left({\mathrm{\&beta;}}^{\&prime;}\right)}{\left(\frac{r}{r_{\max }}\right)}^{{\mathrm{\&alpha;}}^{\&prime;}-1}{(1-\frac{r}{r_{\max }})}^{{\mathrm{\&beta;}}^{\&prime;}-1}---\left(6\right)$

In formula, the form parameter that α ' is this probability density function, the scale parameter that β ' is this probability density function.R and r _maxbe respectively the actual intensity of illumination in this time period and maximum intensity of illumination.

The value of α ' and β ' can draw according to the average value mu of intensity of illumination ' and standard deviation sigma ', and its computing formula is as follows:

${\mathrm{\&alpha;}}^{\&prime;}={\mathrm{\&mu;}}^{\&prime;}(\frac{{\mathrm{\&mu;}}^{\&prime;}(1-{\mathrm{\&mu;}}^{\&prime;})}{{\mathrm{\&sigma;}}^{\&prime;2}}-1)---\left(7\right)$

${\mathrm{\&beta;}}^{\&prime;}={\left(}^{1-\mathrm{\&mu;}\&prime;}(\frac{{\mathrm{\&mu;}}^{\&prime;}(1-{\mathrm{\&mu;}}^{\&prime;})}{{\mathrm{\&sigma;}}^{\&prime;2}}-1)---\left(8\right)$

The photovoltaic battery matrix reduced mechanical model of exerting oneself of gaining merit is:

P _PV＝rAη (9)

A is the gross area of photovoltaic battery matrix in formula (9), and η is the total photoelectric conversion efficiency of photovoltaic battery matrix.

Formula (9) is brought in formula (6) and can show that photovoltaic battery matrix is gained merit the probability density function of exerting oneself.Its expression formula is as follows:

$f\left(P_{\mathrm{PV}}\right)=\frac{\mathrm{\&Gamma;}({\mathrm{\&alpha;}}^{\&prime;}+{\mathrm{\&beta;}}^{\&prime;})}{\mathrm{\&Gamma;}\left({\mathrm{\&alpha;}}^{\&prime;}\right)\mathrm{\&Gamma;}\left({\mathrm{\&beta;}}^{\&prime;}\right)}{\left(\frac{P_{\mathrm{PV}}}{P_{\mathrm{PV},\max }}\right)}^{{\mathrm{\&alpha;}}^{\&prime;}-1}{(1-\frac{P_{\mathrm{PV}}}{P_{\mathrm{PV},\max }})}^{{\mathrm{\&beta;}}^{\&prime;}-1}---\left(10\right)$

As shown in Figure 4, the average value mu ' and standard deviation sigma ' of intensity of illumination is obtained by Monte Carlo simulation.The method utilizes local longitude and latitude in NSDR weather site inquiry monthly average intensity of illumination, generate intensity of illumination per hour, thus the average value mu ' and standard deviation sigma ' of intensity of illumination every day can be obtained, the value of α ' and β ' both can have been drawn according to formula (7) and formula (8).Recycling formula (10), produces at random according to Monte Carlo method setting probability that photovoltaic battery matrix is meritorious exerts oneself.

Meritorious meritorious the exerting oneself with photovoltaic battery matrix of exerting oneself of wind-driven generator provides electric energy to meet power load jointly, when they cannot meet power demands, enable diesel engine generator to generate electricity, and diesel engine needs the time run to need to set according to power load.

Power load presents normal distribution, and its probability density function is as follows:

$f\left(P_L\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&delta;}}_P}\exp [-\frac{{(P_L-{\mathrm{\&mu;}}_P)}^2}{2{\mathrm{\&delta;}}_P^2}]---\left(11\right)$

P in formula _lfor power load power; μ _pand δ _pbe respectively mean value and the standard deviation of power load.

As shown in Figure 5, the mean value of power load and standard deviation can obtain according to Monte Carlo simulation equally, utilize history power load data (as shown in Figure 2) both can obtain its mean value and standard deviation, recycling formula (11), produces the power load result of 1 year 8760h at random according to Monte Carlo method.

The present invention is for the purpose of cost-saving, therefore the cost considering all kinds of distributed power source is needed, but while cost reduces, need to set up a constraints, the present invention is according to the actual conditions of micro-capacitance sensor, constraints is divided into units' installation restricted number, micro-capacitance sensor internal power electric quantity balancing, comprehensive coordination reliability and Environmental Factors.

First cost minimum target function is set up.Micro-capacitance sensor cost comprises unit cost, installation cost and operating cost.For wind, light, bavin micro-capacitance sensor, cost minimization target function C when this micro-capacitance sensor is built _Σ=(N _wG.P _wG.C _wG+ N _pV.P _pV.C _pV+ N _dG.P _dG.C _dG) _min, C in formula _wG, C _pV, C _dGbe respectively the cost of each wind-driven generator, photovoltaic battery matrix, diesel engine generator, N _wG, N _pV, N _dGbe respectively the quantity of wind-driven generator, photovoltaic battery matrix, diesel engine generator.With C ¹, C ², C ³represent unit cost, installation cost and operating cost respectively, then because wind, light micro-capacitance sensor are regenerative resource, operating cost is negligible, and diesel engine generator needs to consume diesel oil, therefore operating cost is needed to take into account, so, cost minimization target function C _Σthen can be expressed as:

$\begin{array}{c}\mathrm{Min}{N}_{\mathrm{WG}}\&CenterDot;(C_{\mathrm{WG}}^1+C_{\mathrm{WG}}^2+C_{\mathrm{WG}}^3)\&CenterDot;P_{\mathrm{WG}}\\ +N_{\mathrm{PV}}\&CenterDot;(C_{\mathrm{PV}}^1+C_{\mathrm{PV}}^2+C_{\mathrm{PV}}^3)\&CenterDot;P_{\mathrm{PV}}\\ +N_{\mathrm{DG}}\&CenterDot;(C_{\mathrm{DG}}^1+C_{\mathrm{DG}}^2+C_{\mathrm{DG}}^3)\&CenterDot;P_{\mathrm{SG}}\end{array}---\left(12\right)$

Wherein be zero.Because micro-capacitance sensor sets up location difference, the generating set that regenerative resource drives also can be different, can only have wind-driven generator, can only have photovoltaic battery matrix, the generating set that also can drive for other renewable energy resources.Should adjust in conjunction with the actual conditions of generating set for formula (12).

The operating cost computing formula of diesel engine generator is as follows:

Cd(P _DG)＝k1+k2·P _DG+k3·P _DG ²(13)

Wherein, k1, k2, k3 are empirical coefficient, and k1 span is 1.0165 ~ 1.1235, k2 span be 0.0624 ~ 0.0690, k3 span is 0.000057 ~ 0.000063.Diesel engine k1, k2, k3 according to different size get different values, and for 30kW diesel engine generator, getting representative value in the present invention is k1=1.0700, k2=0.0657, and k3=0.000060.

Cd (P in formula (13) _dG) be the operating cost of diesel engine generator formula (13) is brought in formula (12) cost minimization target function C can be drawn _Σexpression formula.

Burning due to diesel engine generator can bring the emission problem of carbon dioxide (CO2) simultaneously.Environmental Factors as one of constraints is here carbon dioxide (CO2).CO2 discharge capacity can be expressed as the quadratic function of diesel engine generator power output, as follows:

CO ₂(P _DG)＝a+b·P _DG+c·P _DG ²(14)

Wherein, a, b, c are empirical coefficient, corresponding value is got according to the diesel engine of different model, to get representative value in the present invention be a span is: 0.0267368 ~ 0.0295512, b span is: 0.0016416 ~ 0.0018144, c span is: 0.000001615 ~ 0.000001785.The present embodiment value is: a=0.0281440, b=0.0017280, and c=0.000001700.By setting maximum acceptable CO2 emission limit CO ₂max, can consider CO2 exhaust emission constraint, thus affect the decision-making of configuration scheme in balance of electric power and ener process.。

Micro-capacitance sensor internal power electric quantity balancing, namely the realtime power of micro-capacitance sensor and gross generation will keep balancing with power load:

$N_{\mathrm{WG}}\&CenterDot;P_{\mathrm{WG}.t}+N_{\mathrm{PV}}\&CenterDot;P_{\mathrm{PV}.t}+N_{\mathrm{DG}}\&CenterDot;P_{{\mathrm{DG}}^{.t}}=P_{L.t}---\left(15\right)$

In formula, P _{wG, t}, P _{pV, t}, P _{dG, t}be respectively meritorious when time t of wind-driven generator, photovoltaic battery matrix, diesel engine generator to exert oneself, P _l,tfor power load during t.The selection of t is relevant with the length of the length of project period and planning time yardstick, in the present invention with 1 hour for least unit.

Units' installation quantity can by the restriction in place, and therefore its quantity is in an interval range:

N _i ^min＜＝N _i＜＝N _i ^max(16)

with be respectively all kinds of distributed power source arranging in place the minimum number and maximum quantity that can arrange.Wherein minimum number and maximum quantity can according to the actual conditions in place and economic conditions sets itself.

And the power supply of micro-capacitance sensor needs to have certain reliability, conventional reliability index has: minimum short of electricity probability (LOLP), annual load short of electricity rate LPSP etc., adopt minimum short of electricity probability LOLP to be described, and to set binding occurrence are LOLP in illustrated embodiment of the present invention invention ^max, as follows:

LOLP≤LOLP ^max(17)

Binding occurrence LOLP ^maxsame according to acceptable short of electricity probability sets itself.

Target function and constraints are set up all, then can by hour in units of, with annual 8760h for conceptual phase, Monte Carlo simulation obtain out all kinds of distributed power source per hour go out force data and power load data per hour after, with the installation number of all kinds of distributed power source for decision variable, genetic algorithm is utilized to solve.First decision variable is carried out decimal coded, using target function as fitness function, constraints screening restriction under, by copying, heredity and variation, the evolution and the target function value that finally realize population are minimum.The flow chart of whole algorithm as shown in Figure 8.

Wherein, N _genrepresent population iterations, T represents the time in 1 year, and unit is 1h.The stochastic simulation of wind power generation and photovoltaic generation is unit simulation, and does not consider the correlation of exerting oneself between unit; The difference of distributed power source stochastic simulation and power load stochastic simulation obtains diesel engine ruuning situation, and then obtains the index results such as minimum short of electricity probability LOLP, carbon emission amount.

Can obtain final decision variable by genetic algorithm, the implementation case final decision scheme is in table 1:

Table 1 distributed power source unit number:

The present embodiment case local wind light resources all comparatively horn of plenty, as also obviously found out the complementarity of honourable resource in Fig. 1, is conducive to promoting renewable energy power generation utilance, reduces diesel engine generator running time.

Wherein, the convergence situation of the iterations of algorithm realization and cost as shown in Figure 6.Wherein, micro-capacitance sensor year investment cost change greatly, high power supply reliability can bring high economy cost.When system minimum short of electricity probability LOLP is between 0 ~ 0.015, as shown in Figure 7, when system reliability is set to 0.005, optimum cost is 34,700,000 for minimum short of electricity probability LOLP and Cost sensitivity relation.Along with the increase of minimum short of electricity probability LOLP, the power supply reliability of system is deteriorated, but integrated cost expense also can significantly reduce, and the raising of illustrative system reliability needs to increase investment for cost; And when capacity configuration is enough large, reliable power supply can be ensured completely, but cost now can be very high.In actual micro-grid system capacity configuration, should according to system loading importance, under extreme conditions allow cut-off parts insignificant load during power supply undercapacity, that is according to meeting certain electric power not enough rate restriction requirement, choose optimum power supply capacity combination, thus ensure the economy of system as much as possible.

The above, be only the specific embodiment of the present invention, it should be pointed out that any those of ordinary skill in the art are in the technical scope disclosed by the present invention, the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.

Claims (11)

1. consider an isolated island micro-capacitance sensor capacity configuration optimizing method for randomness, it is characterized in that, comprise the following steps:

In a, analysis micro-capacitance sensor, the relation of all kinds of distributed power source and randomness meteorologic factor, sets up the probabilistic model of the randomness meteorologic factor relevant to all kinds of distributed power source and draws its probability density function;

B, meritoriously exerting oneself and the relational expression of randomness meteorologic factor of calculating all kinds of distributed power source, this relational expression is brought in the probability density function of the randomness meteorologic factor relevant with all kinds of distributed power source, show that all kinds of distributed power source is gained merit the probability density function of exerting oneself;

C, historical statistical data according to the randomness meteorologic factor relevant with all kinds of distributed power source, show that all kinds of distributed power source is gained merit the distributed constant of the probability density function of exerting oneself;

D, to gain merit according to all kinds of the distributed power source probability density function and its distributed constant of exerting oneself, random the meritorious of all kinds of distributed power source that produce is exerted oneself;

E, draw the generated output of micro-capacitance sensor according to meritorious the exerting oneself of all kinds of distributed power source;

F, obtain the probability density function of power load by normal distribution;

G, historical statistical data according to power load, draw the distributed constant of its probability density function;

H, produce power load at random according to the probability density function of power load and the distributed constant of probability density function;

I, based on all kinds of distributed power source, set up cost minimum target function, and using the quantity of the generated output of micro-capacitance sensor and power load Real-time Balancing, all kinds of distributed power source, power supply reliability and Environmental Factors as constraints;

J, with cost minimization target function for fitness function, with the installation number of all kinds of distributed power source for decision variable, the installation number of all kinds of distributed power source when going out cost minimization by genetic algorithm for solving under the screening restriction of constraints.

2. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, is characterized in that: described all kinds of distributed power source comprises diesel engine generator, wind-driven generator and/or photovoltaic battery matrix.

3. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1 or 2, is characterized in that: the randomness meteorologic factor that described and all kinds of distributed power source is relevant comprises wind speed and intensity of illumination.

4. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, is characterized in that: the gain merit distributed constant of the probability density function of exerting oneself, the distributed constant of described power load probability density function of described all kinds of distributed power source is all obtained by Monte Carlo simulation.

5. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, it is characterized in that: in step e, the generated output of micro-capacitance sensor is N _wG.P _wG+ N _pV.P _pV+ N _dG.P _dG, wherein, N _wG, N _pV, N _dGbe respectively the quantity of wind-driven generator, photovoltaic battery matrix, diesel engine generator, P _wG, P _pV, P _dGbe respectively wind-driven generator, photovoltaic battery matrix, the meritorious of diesel engine generator exert oneself.

6. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, is characterized in that: cost minimization target function C in step I _Σ=(N _wG.P _wG.C _wG+ N _pV.P _pV.C _pV+ N _dG.P _dG.C _dG) _min, C in formula _wG, C _pV, C _dGbe respectively the cost of each wind-driven generator, photovoltaic battery matrix, diesel engine generator.

7. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, is characterized in that: in step I, the generated output of micro-capacitance sensor and the Real-time Balancing expression formula of power load power are: $N_{\mathrm{WG}}\&CenterDot;P_{\mathrm{WG}.t}+N_{\mathrm{PV}}\&CenterDot;P_{\mathrm{PV}.t}+N_{\mathrm{DG}}\&CenterDot;P_{{\mathrm{DG}}^{.t}}=P_{L,t},$ In formula, P _{wG, t}, P _{pV, t}, P _{dG, t}be respectively meritorious when time t of wind-driven generator, photovoltaic battery matrix, diesel engine generator to exert oneself, P _l,tfor power load during t.

8. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, it is characterized in that: in step I, the quantity of all kinds of distributed power source is selected according to the scale in place, its span is N _i ^min<=N _i<=N _i ^max, N _i ^minwith N _i ^maxbe respectively all kinds of distributed power source arranging in place the minimum number and maximum quantity that can arrange.

9. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, it is characterized in that: in step I, the judge index of power supply reliability is minimum short of electricity probability LOLP, minimum short of electricity probability LOLP is not more than the maximum short of electricity probable value LOLP of artificial setting ^max.

10. a kind of isolated island micro-capacitance sensor capacity configuration optimizing method considering randomness as claimed in claim 1, it is characterized in that: in step I, Environmental Factors is the discharge capacity of CO2, the computing formula of CO2 discharge capacity is: CO ₂(P _dG)=a+bP _dG+ cP _dG ², in formula, a, b, c are empirical coefficient, by setting the power P that acceptable maximum CO2 discharge capacity is come diesel engine generator _dGlimit.

11. a kind of isolated island micro-capacitance sensor capacity configuration optimizing methods considering randomness as claimed in claim 6, is characterized in that: the cost C of described diesel engine generator _dGcomprise unit cost, installation cost and operating cost, described operating cost Cd (P _dG)=k1+k2P _dG+ k3P _dG ², wherein, k1, k2, k3 are empirical coefficient, and k1 span is 1.0165 ~ 1.1235, k2 span be 0.0624 ~ 0.0690, k3 span is 0.000057 ~ 0.000063.