CN104504263A - Photovoltaic power station harmonic level assessment method based on distribution probability - Google Patents

Abstract

The invention relates to a photovoltaic power station harmonic level assessment method based on distributed probability. The photovoltaic power station harmonic level assessment method based on distributed probability comprises the following steps of fitting a probability distribution curve of different harmonic currents according to testing results of harmonic current in different power sections in a photovoltaic inverter type test; determining features of the harmonic currents of a photovoltaic inverter; and assessing probability distribution of total harmonic current level of a photovoltaic power station according to the probability distribution of the different harmonic currents in the different power sections of the photovoltaic inverter and irradiance distribution of a construction site of the photovoltaic power station so as to achieve the purpose of assessing the quality of electric energy of the power station.

Description

A kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability

Technical field:

The present invention relates to a kind of photovoltaic plant harmonics level appraisal procedure, more specifically relate to a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability.

Background technology:

In the prior art, photovoltaic Power Quality Detection is by photovoltaic cell and photovoltaic DC-to-AC converter, and photo-voltaic power generation station converts sun power to industrial frequency AC electric energy and is transferred on electrical network.Owing to adopting high-frequency power electronic switchgear, the waveform modulated strategy of photovoltaic DC-to-AC converter, topological structure, electrical network background quality of power supply level and operating condition all can have an impact to device grid-connected current harmonic wave size.Along with the increase of photovoltaic plant scale and increasing of input combining inverter number, its grid-connected current harmonic wave pollution to line voltage exported also can not be ignored.Current, power station large-scale photovoltaic power station electric energy quality harmonic test reference standard GB/T/T 14549-93 " quality of power supply utility network harmonic wave ", this standard specifies that the large value of 95% probability is as test result within measuring period, for evaluating the standard of the harmonic pollution order of severity.

Due to photovoltaic plant current harmonic content and irradiance power closely related, the strong and weak situation of test period irradiance has a significant impact test result.For certain 500kW inverter laboratory test results, wherein, test result is the large value of 95% probability of test result under corresponding power interval:

Table 1 inverter A phase current test result

Can see from table 1, this inverter is lower at the interval harmonic content absolute value of low-power, and total harmonic current and rated current ratio THDIn% are 0.94%.Along with output power rises, harmonic content also rises gradually, but when power reaches after more than 70% power, harmonic content no longer rises, and now THDIn% is 2.3%.As can be seen from the above, if when only assessing power station harmonics level with the interval test result of low-power, there is situation about underestimating power station harmonics level.If but only with high power interval test result, power station harmonics level is assessed, consider the singularity that photovoltaic plant runs, the time operated under 100% power interval is also less, and institute is in this way also too harsh. simultaneouslySo the gordian technique of photovoltaic plant Harmonic Assessment be how in conjunction with photovoltaic plant Practical Meteorological Requirements situation in conjunction with each power interval stress_responsive genes result, comprehensive assessment is carried out to photovoltaic plant harmonics level.

Summary of the invention:

The object of this invention is to provide a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability, described method solves the evaluation problem based on the photovoltaic plant harmonic current level of inverter pattern test findings.

For achieving the above object, the present invention is by the following technical solutions: a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability, comprises the following steps:

(1) photovoltaic DC-to-AC converter unit harmonic wave distribution probability is tested;

(2) photovoltaic electric station grid connection point harmonic current level is determined;

(3) the photovoltaic electric station grid connection point harmonic current level of long-term irradiation level is determined.

A kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, the test process of described step (1) comprises the following steps:

Obtain tested photovoltaic DC-to-AC converter reactive power and export Q=0, photovoltaic electric station grid connection point place harmonic current test data;

Determine fundamental current span under photovoltaic plant different capacity, and divide, under described test data is divided into different operating mode according to fundamental current;

Normal distribution is utilized to carry out statistical study to individual harmonic current distribution under different capacity interval;

Obtain the probability density function of photovoltaic DC-to-AC converter several times harmonic wave under different capacity section;

Determine photovoltaic DC-to-AC converter unit harmonic wave distribution probability.

A kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, described tested photovoltaic DC-to-AC converter is connected with photovoltaic array analog DC source; Described photovoltaic DC-to-AC converter and power quality analyzer are all connected with the low-pressure side of many taps step-up transformer; Described many taps step-up transformer is connected with the primary side of described isolating transformer; Control tested photovoltaic DC-to-AC converter reactive power and export Q=0, record harmonic current under 10 power interval continuously by described power quality analyzer.

Another preferred a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, by photovoltaic plant described in the harmonic current distribution tests interpretation of result of described photovoltaic DC-to-AC converter at 1%Pn≤P < 10%Pn, 10%Pn≤P < 20%Pn, 20%Pn≤P < 30%Pn, 30%Pn≤P < 40%Pn, 40%Pn≤P < 50%Pn, 50%Pn≤P < 60%Pn, 60%Pn≤P < 70%Pn, 70%Pn≤P < 80%Pn, harmonic current distribution under 80%Pn≤P < 90%Pn and 90%Pn≤P≤100%Pn different capacity.

A preferred a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability more provided by the invention, supposes that harmonic current Distribution value obeys N (μ, σ ²), wherein μ and σ ²be respectively the unknown parameter of mathematical expectation and variance;

E) likelihood function is

$L(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)=\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\frac{1}{\sqrt{2\mathrm{\&pi;}}\mathrm{\&sigma;}}{e}^{-\frac{{(x_i-\mathrm{\&mu;})}^2}{2{\mathrm{\&sigma;}}^2}}={\left(2\mathrm{\&pi;}{\mathrm{\&sigma;}}^2\right)}^{-\frac{n}{2}}{e}^{-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2}{2{\mathrm{\&sigma;}}^2}}$

F) log-likelihood function is

$l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)=-\frac{n}{2}\ln \left(2\mathrm{\&pi;}{\mathrm{\&sigma;}}^2\right)-\frac{1}{{2\mathrm{\&sigma;}}^2}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2$

G) by l (μ, σ ²) respectively to μ, σ ²ask local derviation, substitute into this primary current different amplitude Frequency statistics result x _i, trying to achieve likelihood equations is:

$\left\{\begin{array}{c}\frac{\&PartialD;l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)}{\&PartialD;\mathrm{\&mu;}}=\frac{1}{{\mathrm{\&sigma;}}^2}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2=0\\ \frac{\&PartialD;l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)}{\&PartialD;{\mathrm{\&sigma;}}^2}=\frac{n}{{2\mathrm{\&sigma;}}^2}+\frac{n}{{2\mathrm{\&sigma;}}^4}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2=0\end{array}\right.$

Solve likelihood equations to obtain

$\left\{\begin{array}{c}\widehat{\mathrm{\&mu;}}=\stackrel{\&OverBar;}{x},\\ {\widehat{\mathrm{\&sigma;}}}^2=\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\stackrel{\&OverBar;}{x})}^2\end{array}\right.$

H) described statistical study is set up all sample observations, therefore replaces observed value x with sample X, obtains μ, σ ²maximum-likelihood estimation be respectively:

$\left\{\begin{array}{c}\widehat{\mathrm{\&mu;}}=\stackrel{\&OverBar;}{X},\\ {\widehat{\mathrm{\&sigma;}}}^2=\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\stackrel{\&OverBar;}{x})}^2=S_n^2\end{array}\right.$

Wherein, n is number of samples; for x _imean value; for sample second-order moment around mean.

Another preferred a kind of described a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, repeats described statistical study step, calculates the distribution of individual harmonic current under all power interval and estimates, obtain respective distribution parameter thus obtain the probability density function of photovoltaic DC-to-AC converter several times harmonic current under different capacity section; P is power interval group; H is the number of times of several times harmonic current.

Another preferably a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, same model photovoltaic DC-to-AC converter harmonic current phase place Normal Distribution, and the equal Normal Distribution of parameter in harmonic current phase place, namely

$y=\mathrm{\&eta;}\sqrt{{\mathrm{\&sigma;}}^2}x+\mathrm{\&eta;}$

Wherein, σ ²be stochastic distribution coefficient (such as 0.01,0.04), η represents test constant, and x is stochastic variable, obeys N (0,1) distribution.

Another preferably a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, described step (2) is by Monte Carlo method determination photovoltaic electric station grid connection point harmonic current level; And independent mutually between each photovoltaic DC-to-AC converter, be independent of each other between individual harmonic current; Photovoltaic DC-to-AC converter harmonic current of the same type and phase place all meet same photovoltaic DC-to-AC converter unit harmonic wave distribution probability;

Carry out random sampling by photovoltaic DC-to-AC converter unit harmonic wave distribution probability and calculate electric station grid connection point harmonic current.

Another preferably a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, described harmonic current deterministic process is:

A) according to this model inverter unit harmonic wave distribution probability random selecting harmonic wave electricity under corresponding power interval

Stream amplitude and phase value;

B) the k subharmonic current that i point injects is I _i,k(θ), total harmonic current of points of common connection is

$I_k\left(\mathrm{\&theta;}\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{I}_{i,k}\left(\mathrm{\&theta;}\right)$

C) repeat above step several times, until also site harmonic current sample meets convergence criterion, represent with coefficient of variation β:

$\mathrm{\&beta;}=\frac{\sqrt{V\left(\hat{E}\left(F\right)\right)}}{\hat{E}\left(F\right)}$

Wherein, $V\left(\hat{E}\left(F\right)\right)=V\left(F\right)/\mathrm{NS},\hat{E}\left(F\right)=\frac{1}{\mathrm{NS}}\underset{i=1}{\overset{\mathrm{NS}}{\mathrm{\&Sigma;}}}F\left(x_i\right),$ NS is total frequency in sampling, x _ithe state sampling of i time, F (x _i) be to i-th sample value x _iexperimental result, V (F) be experiment function F variance.

Another preferably a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability provided by the invention, analyzes the photovoltaic plant their location meteorological record device record historical data of more than a year, obtains solar global irradiance distribution frequency table;

Suppose that irradiance obeys certain distribution H ~ F (x), according to photovoltaic DC-to-AC converter test and the fitting result of probability density function, try to achieve the probability density function I ~ p of photovoltaic DC-to-AC converter each harmonic under certain irradiance interval _y|X(y|x); When considering harmonic current distribution under irradiance distribution and certain power interval, the amplitude distribution of the harmonic current under any condition is a two-dimensional random variable (X, Y), harmonic current distribution probability p in whole irradiance situation _yy () is the marginal probability density function of two-dimensional random variable (X, Y) about harmonic current, determined by following formula:

$p_Y\left(y\right)={\&Integral;}_{-\&infin;}^{+\&infin;}p(x,y)\mathrm{dx}$

Define according to conditional probability density

$P_{Y|X}\left(y\right|x)=\frac{p(x,y)}{p_X\left(x\right)}$

Trying to achieve the probability density function p (x, y) of certain harmonic current under any irradiance is

p(x,y)＝p _Y|X(y|x)p _X(x)

Wherein, p _xx probability density function that () is irradiance distribution, p _y|X(y|x) be harmonic amplitude probability density function under certain appointment irradiance;

By irradiance sliding-model control, determine each distributed area irradiance scope and each interval probability p _x(x); According to the division of different capacity irradiance scope, harmonic current is added up, harmonic amplitude probability density function p in power station under certain irradiance interval _y|X(y|x);

Under discrete conditions, under whole irradiance condition, the probability density function of certain harmonic current is:

$p_Y\left(y\right)=\underset{n=1}{\overset{10}{\mathrm{\&Sigma;}}}{c}_{i}N({\mathrm{\&mu;}}_n,{\mathrm{\&sigma;}}_n)$

According to the definition of accumulated probability density function, certain the subharmonic current probability cumulative distribution function under whole irradiance condition is:

$F_Y\left(y\right)={\&Integral;}_0^y{p}_Y\left(y\right)\mathrm{dy}={\&Integral;}_0^y\underset{n=1}{\overset{10}{\mathrm{\&Sigma;}}}{c}_{n}N({\mathrm{\&mu;}}_n,{\mathrm{\&sigma;}}_n)\mathrm{dy}$

F _yy () is monotonically increasing function, there is inverse function F _y ^-1(y), y ∈ [0,1]; Then consider that the large value of 95% probability of the electric station grid connection point harmonic current of long-term irradiation degree condition is:

$I_{95\%}=F_Y^{-1}\left(0.95\right).$

With immediate prior art ratio, the invention provides technical scheme and there is following excellent effect

1, method of the present invention is by the probability distribution of different subharmonic current under photovoltaic DC-to-AC converter different capacity interval and power plant construction ground irradiation profile feature, the harmonic current injection level of assessment photovoltaic plant;

2, method of the present invention has filled up the blank of current photovoltaic plant harmonic current level evaluation method;

3, method of the present invention is according to the test result of the interval harmonic current of different capacity in photovoltaic DC-to-AC converter type approval test, and the probability distribution curve of the different subharmonic current of matching, determines photovoltaic DC-to-AC converter harmonic current feature;

4, method of the present invention utilizes the probability distribution of photovoltaic DC-to-AC converter different capacity interval, different subharmonic current, the irradiance distribution on ground is built in conjunction with photovoltaic plant, the probability distribution of the overall harmonic current level of assessment photovoltaic plant, reaches the object of the assessment power station quality of power supply;

5, method applicability of the present invention is wider, and dirigibility is strong.

Accompanying drawing explanation

Fig. 1 is photovoltaic DC-to-AC converter electric energy quality test schematic diagram of the present invention;

Fig. 2 is the 10%Pn interval graph of 3 primary current harmonic waves under 500kW inverter different capacity interval of the present invention;

Fig. 3 is the 100%Pn interval graph of 3 primary current harmonic waves under 500kW inverter different capacity interval of the present invention;

Fig. 4 is the interval schematic diagram of 10%Pn of 45 primary current harmonic waves under 500kW inverter different capacity interval of the present invention;

Fig. 5 is the interval schematic diagram of 100%Pn of 45 primary current harmonic waves under 500kW inverter different capacity interval of the present invention;

Fig. 6 is of the present invention and the 10%Pn power interval figure of site 3 subharmonic current;

Fig. 7 is of the present invention and the 100%Pn power interval figure of site 3 subharmonic current;

Fig. 8 is of the present invention and the 10%Pn power interval figure of site 45 subharmonic current;

Fig. 9 is of the present invention and the 100%Pn power interval figure of site 45 subharmonic current;

Figure 10 is certain power station of the present invention annual irradiance data frequency chart; .

Embodiment

Below in conjunction with embodiment, the invention will be described in further detail.

Embodiment 1:

As Figure 1-10 shows, the invention of this example provides a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability, comprises the following steps:

(1) photovoltaic DC-to-AC converter unit harmonic wave distribution probability test

Carry out electric energy quality test to the photovoltaic DC-to-AC converter that photovoltaic plant uses, test topology as shown in fig. 1.The idle output power of inverter is set, controls tested inverter reactive power and export Q=0, record current harmonics under 10 power interval continuously by the power quality analyzer meeting IEC 61000-4-30 ClassA standard.

After obtaining test result, analyze inverter at 1%Pn≤P < 10%Pn, 10%Pn≤P < 20%Pn, 20%Pn≤P < 30%Pn, 30%Pn≤P < 40%Pn, 40%Pn≤P < 50%Pn, 50%Pn≤P < 60%Pn, 60%Pn≤P < 70%Pn, 70%Pn≤P < 80%Pn, harmonic current value under 80%Pn≤P < 90%Pn and 90%Pn≤P≤100%Pn ten power interval.

Utilize normal distribution to carry out statistical study to individual harmonic current distribution under different capacity interval, suppose that harmonic current Distribution value obeys N (μ _,σ ²), wherein μ, σ ²for unknown parameter.

Likelihood function is

$L(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)=\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\frac{1}{\sqrt{2\mathrm{\&pi;}}\mathrm{\&sigma;}}{e}^{-\frac{{(x_i-\mathrm{\&mu;})}^2}{2{\mathrm{\&sigma;}}^2}}={\left(2\mathrm{\&pi;}{\mathrm{\&sigma;}}^2\right)}^{-\frac{n}{2}}{e}^{-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2}{2{\mathrm{\&sigma;}}^2}}$

Log-likelihood function is

$l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)=-\frac{n}{2}\ln \left(2\mathrm{\&pi;}{\mathrm{\&sigma;}}^2\right)-\frac{1}{{2\mathrm{\&sigma;}}^2}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2$

By l (μ, σ ²) respectively to μ, σ ²ask local derviation, substitute into this primary current different amplitude Frequency statistics result x _i, trying to achieve likelihood equations is:

$\left\{\begin{array}{c}\frac{\&PartialD;l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)}{\&PartialD;\mathrm{\&mu;}}=\frac{1}{{\mathrm{\&sigma;}}^2}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2=0\\ \frac{\&PartialD;l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)}{\&PartialD;{\mathrm{\&sigma;}}^2}=\frac{n}{{2\mathrm{\&sigma;}}^2}+\frac{n}{{2\mathrm{\&sigma;}}^4}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2=0\end{array}\right.$

Solve likelihood equations to obtain

$\left\{\begin{array}{c}\widehat{\mathrm{\&mu;}}=\stackrel{\&OverBar;}{x},\\ {\widehat{\mathrm{\&sigma;}}}^2=\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\stackrel{\&OverBar;}{x})}^2\end{array}\right.$

Said process is set up all sample observations, therefore replaces observed value with sample, just obtains μ, σ ²maximum-likelihood estimation be respectively:

$\left\{\begin{array}{c}\widehat{\mathrm{\&mu;}}=\stackrel{\&OverBar;}{X},\\ {\widehat{\mathrm{\&sigma;}}}^2=\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\stackrel{\&OverBar;}{x})}^2=S_n^2\end{array}\right.$

Repeat above step, the distribution calculating each harmonic under other power interval is estimated, obtains respective distribution parameter obtain the probability density function of photovoltaic DC-to-AC converter 3 ~ 50 subharmonic under different capacity section.

Table 2 unit harmonic amplitude probability density function p (μ, σ) parameter

Single inverter harmonic phase angle is determined, can represent with following formula

θ(t)＝α _φh(t)φ _h(t)+β _φh-δ _h，h＝1,3,5,...13

Wherein φ h (t) is background harmonic voltage phase place, and other are coefficients relevant to invertor operation operating mode, and it is as shown in the table for general parameters.

Table 3 harmonic phase distribution parameter

When considering multiple stage current harmonics superposition in photovoltaic plant, also need to consider inverter each harmonic phase problem, when calculating multiple harmonic source current harmonics superposition in power distribution network, consider that harmonic source type is various, duty is various, think that the phase angle of each harmonic is obeyed to be uniformly distributed, θ ~ U [0,2 π].But the photovoltaic DC-to-AC converter switching frequency of same model in photovoltaic plant, control strategy are identical, therefore this method thinks same model photovoltaic DC-to-AC converter harmonic phase Normal Distribution, thinks the equal Normal Distribution of parameter in harmonic phase formula ^[5], namely

$y=\mathrm{\&eta;}\sqrt{{\mathrm{\&sigma;}}^2}x+\mathrm{\&eta;}$

Wherein, σ ²be stochastic distribution coefficient (such as 0.01,0.04), η represents test constant α _h, β φ _hbe stochastic variable with δ h, x, obey N (0,1) distribution.(2) photovoltaic electric station grid connection point current harmonics level calculation

Monte-Carlo Simulation method oneself is known to the stochastic variable of certain probability distribution (probability density function or probability distribution function), obtains the sample data of certain capacity, makes it totally have all identical statistical properties with this distribution.When Monte Carlo method being applied to the assessment of photovoltaic plant harmonics level, each inverter precondition is as follows:

A) independent mutually between each inverter, be independent of each other between each harmonic;

B) inverter harmonic current amplitude of the same type, phase place all meet same photovoltaic DC-to-AC converter unit harmonic wave distribution probability.

On above-mentioned conditioned basic, carrying out random sampling according to inverter unit harmonic wave distribution probability, to calculate electric station grid connection point harmonic current process as follows:

C) according to this model inverter unit harmonic wave distribution probability random selecting harmonic current and phase value under corresponding power interval.

For 500kW inverter, under 10% and 100% power interval, the inverter 3 times of stochastic generation and the distribution of 45 subharmonic currents are as shown in Fig. 2-3 and Fig. 4-5:

In Fig. 2-3,3 subharmonic are 3.22A in the interval harmonic amplitude average of low-power, are 4.11A at high power region average; Phase angle is interval large at the interval variance higher-wattage of low-power, distributes more even.In Fig. 3,45 subharmonic are 6.1A in the interval harmonic amplitude average of low-power, are 6.2A at high power region average; The interval harmonic amplitude of low-power and PHASE DISTRIBUTION basically identical.

D) the k subharmonic current that i point injects is I _i,k(θ), total harmonic current of points of common connection is

$I_k\left(\mathrm{\&theta;}\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{I}_{i,k}\left(\mathrm{\&theta;}\right)$

C) repeat above step M time (M is a number greatly), until also site harmonic amplitude sample meets convergence criterion, usually represent with coefficient of variation β:

$\mathrm{\&beta;}=\frac{\sqrt{V\left(\hat{E}\left(F\right)\right)}}{\hat{E}\left(F\right)}$

Wherein, $V\left(\hat{E}\left(F\right)\right)=V\left(F\right)/\mathrm{NS},\hat{E}\left(F\right)=\frac{1}{\mathrm{NS}}\underset{i=1}{\overset{\mathrm{NS}}{\mathrm{\&Sigma;}}}F\left(x_i\right),$ NS is total frequency in sampling, x _ithe state sampling of i time, F (x _i) be to i-th sample value x _iexperimental result, V (F) be experiment function F variance.

For a certain 50MW power station, grid-connected point voltage is 35kV, containing 500kW inverter routine on 100 platforms, then and the assessment result of 3 times, site and 45 subharmonic as shown in Fig. 6-7 and Fig. 8-9:

Utilize unit probability density function acquiring method in step 1, through parameter estimation and inspection, obtain the current amplitude distribution probability function of power station under different capacity interval.As shown in table 4

Table 4 electric station grid connection point harmonic wave probability density function P (μ, σ) parameter

(3) the photovoltaic electric station grid connection point current harmonics level calculation of long-term irradiation level is considered

Carry out statistical study to their location, power station long history irradiance data, obtain irradiance distribution frequency table as shown in Figure 10, wherein transverse axis is that irradiance is strong and weak, and unit is W/m ².The longitudinal axis is the irradiance frequency being in this interval.

Suppose that irradiance obeys certain distribution H ~ F (x), according to inverter test and the fitting result of probability density function, can in the hope of the probability density function I ~ p of inverter each harmonic under certain irradiance interval _y|X(y|x).When considering harmonic current distribution under irradiance distribution and certain power interval, the amplitude distribution of the harmonic current under any condition is a two-dimensional random variable (X, Y), harmonic current distribution probability p in whole irradiance situation _yy () is the marginal probability density function of two-dimensional random variable (X, Y) about harmonic current, be shown below:

$p_Y\left(y\right)={\&Integral;}_{-\&infin;}^{+\&infin;}p(x,y)\mathrm{dx}$

Define according to conditional probability density

$P_{Y|X}\left(y\right|x)=\frac{p(x,y)}{p_X\left(x\right)}$

Trying to achieve the probability density function p (x, y) of certain harmonic current under any irradiance is

p(x,y)＝p _Y|X(y|x)p _X(x)

Wherein, p _xx probability density function that () is irradiance distribution, p _y|X(y|x) be harmonic amplitude probability density function under certain appointment irradiance.

Consider that the irradiance distribution under long-term behavior disobeys standard profile form, cause calculated amount excessive, by irradiance sliding-model control, each distributed area irradiance scope and each interval probability p _xx () is as shown in the table:

The interval irradiance scope of table 6 different capacity

Interval	1	2	…	10
					Irradiance interval (W/m 2)	0<x≤100	100<x≤200	…	900<x≤1000
Probability p X(x)	c 1	c 2		c 10

According to the division of different capacity irradiance scope, harmonic current is added up, harmonic amplitude probability density function p in power station under certain irradiance interval _y|X(y|x) can reference table 4.For 3 subharmonic, under different irradiance, harmonic current distribution probability p (y, x) is as shown in table 7:

Harmonic amplitude probability distribution under table 73 subharmonic different capacity interval

Under discrete conditions, under whole irradiance condition, the probability density function of certain harmonic current is:

$p_Y\left(y\right)=\underset{n=1}{\overset{10}{\mathrm{\&Sigma;}}}{c}_{i}N({\mathrm{\&mu;}}_n,{\mathrm{\&sigma;}}_n)$

According to the definition of accumulated probability density function, the expression formula of certain the subharmonic current probability cumulative distribution function under whole irradiance condition is:

$F_Y\left(y\right)={\&Integral;}_0^y{p}_Y\left(y\right)\mathrm{dy}={\&Integral;}_0^y\underset{n=1}{\overset{10}{\mathrm{\&Sigma;}}}{c}_{n}N({\mathrm{\&mu;}}_n,{\mathrm{\&sigma;}}_n)\mathrm{dy}$

Obviously, F _yy () is monotonically increasing function, there is inverse function F _y ^-1(y), y ∈ [0,1].Then consider that the large value of 95% probability of the electric station grid connection point harmonic current of long-term irradiation degree condition is:

$I_{95\%}=F_Y^{-1}\left(0.95\right)$

Finally should be noted that: above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit; although those of ordinary skill in the field are to be understood that with reference to above-described embodiment: still can modify to the specific embodiment of the present invention or equivalent replacement; these do not depart from any amendment of spirit and scope of the invention or equivalent replacement, are all applying within the claims of the present invention awaited the reply.

Claims (10)

1., based on a photovoltaic plant harmonics level appraisal procedure for distribution probability, it is characterized in that: comprise the following steps:

(1) photovoltaic DC-to-AC converter unit harmonic wave distribution probability is tested;

(2) photovoltaic electric station grid connection point harmonic current level is determined;

(3) the photovoltaic electric station grid connection point harmonic current level of long-term irradiation level is determined.

2. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 1, is characterized in that: the test process of described step (1) comprises the following steps:

Obtain tested photovoltaic DC-to-AC converter reactive power and export Q=0, photovoltaic electric station grid connection point place harmonic current test data;

Determine fundamental current span under photovoltaic plant different capacity, and divide, under described test data is divided into different operating mode according to fundamental current;

Normal distribution is utilized to carry out statistical study to individual harmonic current distribution under different capacity interval;

Obtain the probability density function of photovoltaic DC-to-AC converter several times harmonic wave under different capacity section;

Determine photovoltaic DC-to-AC converter unit harmonic wave distribution probability.

3. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 2, is characterized in that: described tested photovoltaic DC-to-AC converter is connected with photovoltaic array analog DC source; Described photovoltaic DC-to-AC converter and power quality analyzer are all connected with the low-pressure side of many taps step-up transformer; Described many taps step-up transformer is connected with the primary side of described isolating transformer; Control tested photovoltaic DC-to-AC converter reactive power and export Q=0, record harmonic current under 10 power interval continuously by described power quality analyzer.

4. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 2, it is characterized in that: by photovoltaic plant described in the harmonic current distribution tests interpretation of result of described photovoltaic DC-to-AC converter at 1%Pn≤P < 10%Pn, 10%Pn≤P < 20%Pn, 20%Pn≤P < 30%Pn, 30%Pn≤P < 40%Pn, 40%Pn≤P < 50%Pn, 50%Pn≤P < 60%Pn, 60%Pn≤P < 70%Pn, 70%Pn≤P < 80%Pn, harmonic current distribution under 80%Pn≤P < 90%Pn and 90%Pn≤P≤100%Pn different capacity.

5. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 3, is characterized in that: suppose that harmonic current Distribution value obeys N (μ, σ ²), wherein μ and σ ²be respectively the unknown parameter of mathematical expectation and variance;

A) likelihood function is

$L(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)=\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\frac{1}{\sqrt{2\mathrm{\&pi;}}\mathrm{\&sigma;}}{e}^{-\frac{{(x_i-\mathrm{\&mu;})}^2}{2{\mathrm{\&sigma;}}^2}}={\left(2\mathrm{\&pi;}{\mathrm{\&sigma;}}^2\right)}^{-\frac{n}{2}}{e}^{-\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2}{2{\mathrm{\&sigma;}}^2}}$

B) log-likelihood function is

$l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)=-\frac{n}{2}\ln \left(2\mathrm{\&pi;}{\mathrm{\&sigma;}}^2\right)-\frac{1}{2{\mathrm{\&sigma;}}^2}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2$

C) by l (μ, σ ²) respectively to μ, σ ²ask local derviation, substitute into this primary current different amplitude Frequency statistics result x _i, trying to achieve likelihood equations is:

$\left\{\begin{array}{c}\frac{\&PartialD;l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)}{\&PartialD;\mathrm{\&mu;}}=\frac{1}{{\mathrm{\&sigma;}}^2}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2=0\\ \frac{\&PartialD;l(\mathrm{\&mu;},{\mathrm{\&sigma;}}^2)}{\&PartialD;{\mathrm{\&sigma;}}^2}=\frac{n}{2{\mathrm{\&sigma;}}^2}+\frac{n}{2{\mathrm{\&sigma;}}^4}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\mathrm{\&mu;})}^2=0\end{array}\right.$

Solve likelihood equations to obtain

$\left\{\begin{array}{c}\widehat{\mathrm{\&mu;}}=\stackrel{\&OverBar;}{x}\\ {\widehat{\mathrm{\&sigma;}}}^2=\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\stackrel{\&OverBar;}{x})}^2\end{array}\right.,$

D) described statistical study is set up all sample observations, therefore replaces observed value x with sample X, obtains μ, σ ²maximum-likelihood estimation be respectively:

$\left\{\begin{array}{c}\widehat{\mathrm{\&mu;}}=\stackrel{\&OverBar;}{x}\\ {\widehat{\mathrm{\&sigma;}}}^2=\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{(x_i-\stackrel{\&OverBar;}{x})}^2=S_n^2\end{array}\right.,$

Wherein, n is number of samples; for sample mean; for sample second-order moment around mean.

6. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as described in claim 3-5 any one, it is characterized in that: repeat described statistical study step, the distribution calculating individual harmonic current under all power interval is estimated, obtains respective distribution parameter thus obtain the probability density function of photovoltaic DC-to-AC converter several times harmonic current under different capacity section; P is power interval group; H is the number of times of several times harmonic current.

7. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 6, is characterized in that: same model photovoltaic DC-to-AC converter harmonic current phase place Normal Distribution, and the equal Normal Distribution of parameter in harmonic current phase place, namely

$y=\mathrm{\&eta;}\sqrt{{\mathrm{\&sigma;}}^2}x+\mathrm{\&eta;}$

Wherein, σ ²be stochastic distribution coefficient (such as 0.01,0.04), η represents test constant, and x is stochastic variable, obeys N (0,1) distribution.

8. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 1, is characterized in that: described step (2) is by Monte Carlo method determination photovoltaic electric station grid connection point harmonic current level; And independent mutually between each photovoltaic DC-to-AC converter, be independent of each other between individual harmonic current; Photovoltaic DC-to-AC converter harmonic current of the same type and phase place all meet same photovoltaic DC-to-AC converter unit harmonic wave distribution probability;

Carry out random sampling by photovoltaic DC-to-AC converter unit harmonic wave distribution probability and calculate electric station grid connection point harmonic current.

9. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 8, is characterized in that: described harmonic current deterministic process is:

A) according to this model inverter unit harmonic wave distribution probability random selecting harmonic current and phase value under corresponding power interval;

B) the k subharmonic current that i point injects is I _i,k(θ), total harmonic current of points of common connection is

$I_k\left(\mathrm{\&theta;}\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{I}_{i,k}\left(\mathrm{\&theta;}\right)$

C) repeat above step several times, until also site harmonic current sample meets convergence criterion, represent with coefficient of variation β:

$\mathrm{\&beta;}=\frac{\sqrt{V\left(\hat{E}\left(F\right)\right)}}{\hat{E}\left(F\right)}$

Wherein, $V\left(\hat{E}\left(F\right)\right)=V\left(F\right)/\mathrm{NS},\hat{E}\left(F\right)=\frac{1}{\mathrm{NS}}\underset{i=1}{\overset{\mathrm{NS}}{\mathrm{\&Sigma;}}}F\left(x_i\right),$ NS is total frequency in sampling, x _ithe state sampling of i time, F (x _i) be to i-th sample value x _iexperimental result, V (F) be experiment function F variance.

10. a kind of photovoltaic plant harmonics level appraisal procedure based on distribution probability as claimed in claim 1, it is characterized in that: utilize the meteorological record device record historical data of more than a year to analyze to photovoltaic plant their location, obtain solar global irradiance distribution frequency table;

Suppose that irradiance obeys certain distribution H ~ F (x), according to photovoltaic DC-to-AC converter test and the fitting result of probability density function, try to achieve the probability density function I ~ p of photovoltaic DC-to-AC converter each harmonic under certain irradiance interval _y|X(y|x); When considering harmonic current distribution under irradiance distribution and certain power interval, the amplitude distribution of the harmonic current under any condition is a two-dimensional random variable (X, Y), in whole irradiance situation, harmonic current distribution probability pY (y) is two-dimensional random variable (X, Y) about the marginal probability density function of harmonic current, determined by following formula:

$p_Y\left(y\right)={\&Integral;}_{-\&infin;}^{+\&infin;}p(x,y)\mathrm{dx}$

Define according to conditional probability density

$p_{Y|X}\left(y\right|x)=\frac{p(x,y)}{p_X\left(x\right)}$

Trying to achieve the probability density function p (x, y) of certain harmonic current under any irradiance is

p(x,y)＝p _Y|X(y|x)p _X(x)

Wherein, p _xx probability density function that () is irradiance distribution, p _y|X(y|x) be harmonic amplitude probability density function under certain appointment irradiance;

By irradiance sliding-model control, determine each distributed area irradiance scope and each interval probability p _x(x); According to the division of different capacity irradiance scope, harmonic current is added up, harmonic amplitude probability density function p in power station under certain irradiance interval _y|X(y|x);

Under discrete conditions, under whole irradiance condition, the probability density function of certain harmonic current is:

$p_Y\left(y\right)=\underset{n=1}{\overset{10}{\mathrm{\&Sigma;}}}{c}_{i}N({\mathrm{\&mu;}}_n,{\mathrm{\&sigma;}}_n)$

According to the definition of accumulated probability density function, certain the subharmonic current probability cumulative distribution function under whole irradiance condition is:

$F_Y\left(y\right)={\&Integral;}_0^y{p}_Y\left(y\right)\mathrm{dy}={\&Integral;}_0^y\underset{n=1}{\overset{10}{\mathrm{\&Sigma;}}}{c}_{n}N({\mathrm{\&mu;}}_n,{\mathrm{\&sigma;}}_n)\mathrm{dy}$

F _yy () is monotonically increasing function, there is inverse function y ∈ [0,1]; Then consider that the large value of 95% probability of the electric station grid connection point harmonic current of long-term irradiation degree condition is:

$I_{95\%}=F_Y^{-1}\left(0.95\right).$