CN104504618A - Micro-grid reliability evaluation data sampling method based on pair-copula function - Google Patents

Abstract

The invention discloses a micro-grid reliability evaluation data sampling method based on pair-copula function; A Spearsman rank parameter method is used for statistically analyzing and processing wind speed, illumination intensity and load data which are collected in a micro-grid, so as to obtain random correlation parameters of the wind speed, the illumination intensity and the load data inside the micro-grid; the random correlation parameters are introduced into the micro-grid reliability evaluation algorithm wind speed, illumination intensity and load data sampling process based on the pair-copula function; data of the wind speed, illumination intensity and the hours of load is sampled by a function F (x1, x2, x3); the sampled data is used in the micro-grid reliability evaluation algorithm; the real time output power of the fan and photovoltaic device in the micro-grid is calculated by the relationship between the wind speed and fan output power, and the relationship between the illumination intensity and the photovoltaic device, so as to obtain the sample data based on the pair-copula function. The micro-grid reliability evaluation data sampling method based on pair-copula function solves the problem that the random correlation of the wind speed, the illumination intensity and the load data cannot be considered in the conventional probabilistic model; the precision of sample data in the micro-grid reliability evaluation algorithm is increased.

Description

Based on the micro-capacitance sensor reliability assessment sampling of data method of pair-copula function

Technical field

The present invention relates to micro-capacitance sensor assessment technology, particularly relate to a kind of micro-capacitance sensor reliability assessment sampling of data method based on pair-copula function.

Background technology

In micro-capacitance sensor, exerting oneself of distributed power source has uncertainty, in order to assess the reliability of micro-capacitance sensor, need to set up the probability model of distributed power source power producing characteristics and part throttle characteristics in micro-capacitance sensor, and sample wind speed, intensity of illumination and load data to assess the reliability of micro-capacitance sensor based on the probability model set up by the methods of sampling.In existing micro-capacitance sensor Reliability Evaluation Model, often suppose that in micro-capacitance sensor, the different distributions formula energy is separate between exerting oneself.In fact, existing document [1-3] proves, in micro-capacitance sensor, the primary energy of distributed power source comprises the load in wind speed, intensity of illumination and microgrid, there is random correlativity between them, the random correlativity ignoring wind speed in micro-capacitance sensor, intensity of illumination and load will affect the accuracy of micro-capacitance sensor reliability assessment.

In current description microgrid reliability assessment, between variable, the method for random correlativity mainly contains: joint probability density function method, the Correlation Moment tactical deployment of troops and copula function method etc.For obeying the variable of same marginal distribution, by joint probability density function method, correlativity between variable can be described; For disobeying the variable of same distribution, the Correlation Moment tactical deployment of troops and copula function method can describe the correlativity between variable.But the Correlation Moment tactical deployment of troops and copula function method are difficult to the problem describing more than ternary.Pair-copula function, as a kind of method describing random correlativity between multiple variable, can describe the random correlativity between multiple variable easily.

When assessing micro-capacitance sensor power supply reliability, the quality of data from the sample survey directly affects the accuracy of micro-capacitance sensor Reliability Evaluation result.For wind speed in micro-capacitance sensor, random correlativity between intensity of illumination and load, pair-copula functional based method is incorporated in the sampling of data process of micro-capacitance sensor reliability assessment by the present invention, establishes the micro-capacitance sensor reliability assessment sampling of data method based on pair-copula function.

[1]Zhilong Qin,Wenyuan Li,Xiaofu Xiong.Incorporating multiple correlations amongwind speeds,photovoltaic powers and bus loads in composite system reliability evaluation.ApplyEnergy,2013,110:285-294.

[2]George Papaefthymiou,Dorota Kurowicka.Using copula for modeling stochasticdependenceinpowersystemuncertaintyanalysis.IEEETransactiononPower Systems,2009,24(1):40-49.

[3]H.ValizadehHaghi M,Tavakoli Bina,MAGolker,S MMoghaddas.Using copulasfor analysis of large datasets in renewable distribution generation:PV and wind powerintegrationinIran.RenewableEnergy,2010,35(9):1991-2000.

Summary of the invention

In order to overcome above-mentioned prior art Problems existing, the present invention proposes a kind of micro-capacitance sensor reliability assessment sampling of data method based on pair-copula function, Spearsman order parametric technique is utilized to obtain the random relevance parameter of wind speed in micro-capacitance sensor, intensity of illumination and load statistics, random relevance parameter is brought in the sampling process based on the micro-capacitance sensor Reliability Evaluation Algorithm wind speed of pair-copula function, intensity of illumination and load data, solve the problem cannot considering the random correlativity of wind speed, intensity of illumination and load data in conventional probability model; Improve the accuracy of data from the sample survey in microgrid Reliability Evaluation Algorithm.Carry out statistical study process

The present invention proposes a kind of micro-capacitance sensor reliability assessment sampling of data method based on pair-copula functional based method, the method comprises the following steps:

The annual data of the wind speed collected every 1 hour in step 1, input micro-capacitance sensor, intensity of illumination, load;

Step 2, utilization Spearman order parametric technique calculate the random rank correlation parameter of wind speed, intensity of illumination and load in micro-capacitance sensor, utilize the Spearman order parameter ρ between wind speed and intensity of illumination, between wind speed and load, between intensity of illumination and load statistics ₁, ρ ₂, ρ ₃, obtaining random relevance parameter ρ is ρ=(ρ ₁, ρ ₂, ρ ₃);

Step 3, statistical property according to wind speed, intensity of illumination and load, the marginal distribution of calculation of wind speed, intensity of illumination and load statistics;

Step 4, the random relevance parameter ρ drawn by Spearsman order parametric technique substitute into pair-copula function, obtain the random dependency expression formula of wind speed in micro-capacitance sensor, intensity of illumination and load:

F(x ₁,x ₂,x ₃)＝C(F ₁(x ₁),F ₂(x ₂),F ₃(x ₃),ρ) (1)

In formula: x ₁, x ₂, x ₃represent wind speed, intensity of illumination and load variation respectively; F ₁(x ₁), F ₂(x ₂), F ₃(x ₃) be respectively variable x ₁, x ₂, x ₃marginal distribution; ρ is the Spearsman rank correlation parameter calculated in step (2); C is pair-copula function;

Step 5, by function F (x ₁, x ₂, x ₃) hour data of sampling wind speed, intensity of illumination and load, in micro-capacitance sensor Reliability Evaluation Algorithm, calculate exerting oneself in real time of blower fan and photovoltaic apparatus in micro-capacitance sensor by wind speed and blower fan relation, intensity of illumination and photovoltaic relation of exerting oneself of exerting oneself; Sampling process comprises the following steps:

Uniform random number w is produced in interval [0,1] ₁, w ₂, w ₃;

Make x ₁=w ₁; By w ₂bring formula x into ₂=h ^-1(w ₂, x ₁, ρ ₁), obtain x ₂;

By w ₂, w ₃bring formula x into ₃=h ^-1[h ^-1(w ₃, x ₂, ρ ₃), x ₁, p ₂], obtain x ₃;

(x ₁, x ₂, x ₃) be the random number met the demands, substitute into the marginal distribution function F of wind speed, intensity of illumination, load ₁(x ₁), F ₂(x ₂), F ₃(x ₃) namely obtain the sample value of wind speed, intensity of illumination and load; Function F (x ₁, x ₂, x ₃) by its inverse function h _- ¹(x, y, ρ) calculates:

$h^{-1}(x,y,\mathrm{\ρ})=\mathrm{\φ}({\mathrm{\φ}}^{-1}\left(x\right)\·\sqrt{1-{\mathrm{\ρ}}^2}+\mathrm{\ρ}\·{\mathrm{\φ}}^{-1}\left(y\right));$

Wherein:

${\mathrm{\φ}}^{-1}\left(x\right)=\sqrt{2}{\mathrm{erf}}^{-1}(2x-1);$

${\mathrm{erf}}^{-1}\left(x\right)=\mathrm{sgn}\left(x\right)\·\sqrt{\sqrt{{(\frac{2}{\mathrm{\π}\·a}+\frac{\ln (1-x^2)}{2})}^2-\frac{\ln (1-x^2)}{a}}-(\frac{2}{\mathrm{\π}\·a}+\frac{\ln (1-x^2)}{2})};$

a＝0.147；

So far the data from the sample survey based on pair-copula function is obtained, for calculating the reliability of micro-capacitance sensor.

Compared with prior art, the present invention makes up the deficiency of data modeling and the methods of sampling in traditional microgrid reliability estimation method, describe the randomness of wind speed in micro-capacitance sensor, intensity of illumination and load long-time statistical data and random correlativity better, propose the micro-capacitance sensor reliability assessment sampling of data method based on pair-copula functional based method.

Accompanying drawing explanation

Fig. 1 is overall flow figure of the present invention.

Embodiment

Below in conjunction with accompanying drawing, technical solution of the present invention and embodiment are further described.

The object of the invention is the deficiency making up sampling of data method in traditional micro-capacitance sensor reliability estimation method, describe the randomness of variable in micro-capacitance sensor reliability assessment and random correlativity better, propose the micro-capacitance sensor reliability assessment sampling of data method based on pair-copula functional based method.

Preferred forms

The annual data of the wind speed collected every 1 hour in input micro-capacitance sensor, intensity of illumination, load;

Spearman order parametric technique is used to calculate micro-capacitance sensor wind speed, random relevance parameter between intensity of illumination and load; According to the statistical property of wind speed, intensity of illumination and load, the parameter of the marginal distribution of calculation of wind speed, intensity of illumination and load;

Bring the random relevance parameter that Spearsman order parametric technique draws into pair-copula function, by the hour data of the above random sampling of dependency expression formula wind speed, intensity of illumination and load based on pair-copula function, for in the many micro-grid systems Reliability Evaluation Algorithm of region, to be exerted oneself exerting oneself in real time of blower fan and photovoltaic apparatus in the many micro-grid systems in relation zoning by exert oneself relation, intensity of illumination and photovoltaic of wind speed and blower fan; So far obtain the data from the sample survey based on pair-copula function, can be used for the reliability of the many micro-grid systems in zoning.

Further illustrate the application of sampling of data method in microgrid reliability assessment based on pair-copula function in the present invention for IEEE RBTS-BUS 2 example, wherein micro-capacitance sensor is connected in feeder line 1 Circuit Fault on Secondary Transformer being numbered 7.Set up the wind speed of pair-copula function needs, intensity of illumination and load data from nascent state city actual measurement data in Tianjin.

Table 1 is the rated power data of relevant device in micro-capacitance sensor.

Table 2 is the random correlativity calculation result of micro-capacitance sensor apoplexy, light and load.

Table 3 is the reliability assessment result under micro-capacitance sensor two kinds of scenes.

Table 1

Table 2

Data type	Wind	Light	Load
				Wind	1	0.1810	0.3172
Light	0.1810	1	0.2307
				Load	0.3172	0.2307	1

Table 3

Result of calculation shows, what the present invention proposed can consider wind speed in microgrid, random correlativity between intensity of illumination and load based on the data from the sample survey method in the micro-capacitance sensor Reliability Evaluation Algorithm of the pair-copula function data methods of sampling, and classic method cannot consider this random correlativity.

Claims (1)

1., based on a micro-capacitance sensor reliability assessment sampling of data method for pair-copula functional based method, it is characterized in that, the method comprises the following steps:

The annual data of the wind speed collected every 1 hour in step (1), input micro-capacitance sensor, intensity of illumination, load;

Step (2), utilization Spearman order parametric technique calculate the random rank correlation parameter of wind speed, intensity of illumination and load in micro-capacitance sensor, utilize the Spearman order parameter ρ between wind speed and intensity of illumination, between wind speed and load, between intensity of illumination and load statistics ₁, ρ ₂, ρ ₃, obtaining random relevance parameter ρ is ρ=(ρ ₁, ρ ₂, ρ ₃);

Step (3), statistical property according to wind speed, intensity of illumination and load, the marginal distribution of calculation of wind speed, intensity of illumination and load statistics;

Step (4), the random relevance parameter ρ drawn by Spearsman order parametric technique substitute into pair-copula function, obtain the random dependency expression formula of wind speed in micro-capacitance sensor, intensity of illumination and load:

F(x ₁,x ₂,x ₃)＝C(F ₁(x ₁),F ₂(x ₂),F ₃(x ₃),ρ) (1)

In formula: x ₁, x ₂, x ₃represent wind speed, intensity of illumination and load variation respectively; F ₁(x ₁), F ₂(x ₂), F ₃(x ₃) be respectively variable x ₁, x ₂, x ₃marginal distribution; ρ is the Spearsman rank correlation parameter calculated in step (2); C is pair-copula function;

Step (5), by function F (x ₁, x ₂, x ₃) hour data of sampling wind speed, intensity of illumination and load, in micro-capacitance sensor Reliability Evaluation Algorithm, calculate exerting oneself in real time of blower fan and photovoltaic apparatus in micro-capacitance sensor by wind speed and blower fan relation, intensity of illumination and photovoltaic relation of exerting oneself of exerting oneself; Sampling process comprises the following steps:

Uniform random number w is produced in interval [0,1] ₁, w ₂, w ₃;

Make x ₁=w ₁; By w ₂bring formula x into ₂=h ^-1(w ₂, x ₁, ρ ₁), obtain x ₂;

By w ₂, w ₃bring formula x into ₃=h ^-1[h ^-1(w ₃, x ₂, ρ ₃), x ₁, p ₂], obtain x ₃;

(x ₁, x ₂, x ₃) be the random number met the demands, substitute into the marginal distribution function F of wind speed, intensity of illumination, load ₁(x ₁), F ₂(x ₂), F ₃(x ₃) namely obtain the sample value of wind speed, intensity of illumination and load; Function F (x ₁, x ₂, x ₃) by its inverse function h _- ¹(x, y, ρ) calculates:

Wherein:

${\mathrm{\φ}}^{-1}\left(x\right)=\sqrt{2}{\mathrm{erf}}^{-1}(2x-1);$

${\mathrm{erf}}^{-1}\left(x\right)=\mathrm{sgn}\left(x\right)\·\sqrt{\sqrt{(\frac{2}{\mathrm{\π}\·a}+\frac{\ln (1-x^2)}{2}{)}^2-\frac{\ln (1-x^2)}{a}}-(\frac{2}{\mathrm{\π}\·a}+\frac{\ln (1-x^2)}{2})};$

a＝0.147；

So far the data from the sample survey based on pair-copula function is obtained, for calculating the reliability of micro-capacitance sensor.