CN102594216B - Probability assessment method of effects of distributed photovoltaic power supply access - Google Patents

Abstract

A probability assessment method of effects of distributed photovoltaic power supply access mainly includes inputting initial data including network line parameter, generator power and load node power, type and nominal power of a distributed power supply and the like; inputting observation light intensity of access places of the distributed photovoltaic power supply at 24 hours one day; calculating output power PM initial value of the distributed photovoltaic power supply; calculating power flow distribution under normal operation condition; calculating each-stage cumulant of random variable of output power of the distributed photovoltaic power supply; calculating each-stage cumulant deltas<(k)> of node random injection power according to property of the cumulant; further calculating each-stage cumulant deltaX<(k)> of state variable; and calculating system voltage out-of-limit probability of one year to assess effects of distributed photovoltaic power supply access. The probability assessment method completely considers load power and probability distribution characteristics of the distributed photovoltaic power supply output and is capable of accurately assessing effects of distributed power supply access. The method is suitable for probability assessment of effects of distributed photovoltaic power supply access.

Description

A kind of probability assessment method of distributed photovoltaic power access impact

Technical field

A kind of probability assessment method that the present invention relates to distributed photovoltaic power access impact, belongs to photo-voltaic power supply interconnection technology field.

Background technology

The mode that large electrical network combines with distributed power source (DG) is the important way that can reduce investment outlay, reduce energy consumption, improve power system reliability and flexibility, is also the main development direction of following electrical network.Although distributed power source access can bring a series of benefit to operation of power networks, but the access of a large amount of distributed power sources can have a huge impact the structure of distribution and operation too, its the underlying cause is that after distributed power source access, distribution trend will be no longer one way flow, radioactive network spreads all over power supply and the interconnected network of user by becoming, trend also not necessarily uniaxially from substation bus bar, flow to each load, likely there will be and reflux and complicated change in voltage.

As a kind of typical types in distributed power source, meritorious the exerting oneself of distributed photovoltaic power is subject to the impact of natural weather condition very large, and it is exerted oneself and changes along with the variation of light intensity.Like this, while containing a large amount of distributed photovoltaic powers in system, its meritorious unsteadiness of exerting oneself can cause the sudden change of system load flow, and the impact that now only adopts traditional tidal current computing method to assess distributed photovoltaic power access will not be most suitable.

Summary of the invention

The object of the invention is, in order to take into full account fluctuation and the randomness of load power and distributed photovoltaic power output, the impact that accurate evaluation distributed photovoltaic power grid causes, the invention provides a kind of probability assessment method that distributed photovoltaic power access affects.

Technical scheme of the present invention is that the present invention adopts following steps to implement probability assessment:

1) initial data such as the type of input network line parameter, generator and load bus power and distributed power source, rated power;

2) the input distributed photovoltaic power access place observation intensity of illumination of a day 24 hours;

3) by formula P _mthe power output P of=rA η Computation distribution formula photo-voltaic power supply _minitial value, wherein A is the square formation gross area, η is the total photoelectric conversion efficiency of square formation, the intensity of illumination that r is certain area;

4) apply the trend distribution that traditional tidal current computing method calculates normal operation, obtain the state variable X on benchmark operating point ₀, Jacobian matrix J ₀;

5) each rank cumulant of Computation distribution formula photo-voltaic power supply power output stochastic variable;

6) according to each rank cumulant Δ S of the random injecting power of character computing node of cumulant ^(k);

7) by injecting at random each rank cumulant Δ S ^(k)obtain each rank cumulant Δ X of state variable ^(k);

8) by distribution function and the probability density function of Gram-Charlier progression computing mode variable Δ X;

9) calculate i days N probability P that node voltage is out-of-limit _k(k=1,2 ... N), ask for the probability P of this day total voltage out-of-limit _i:

10) repeat second to the 9th step, calculate the out-of-limit probability of system voltage every day in a year, year dangerous hourage of expectation voltage can be obtained by following formula:

The power output P of distributed photovoltaic power in the present invention _mprobability density function f (P _m) become Beta to distribute, in formula, α and β are the form parameters that Beta distributes, and Γ is Gamma function, R _mfor square formation peak power output.

The present invention adopts formula calculate the v rank square α of stochastic variable x _v, wherein f (x) is the probability density function of stochastic variable x.For simplifying, calculate, cumulant is only got front 6 rank, by each rank square, adopts following formula to ask for front 6 rank cumulant:

γ ₁＝α ₁＝m

${\mathrm{\&gamma;}}_2={\mathrm{\&alpha;}}_2-{\mathrm{\&alpha;}}_1^2$

${\mathrm{\&gamma;}}_3={\mathrm{\&alpha;}}_3-3{\mathrm{\&alpha;}}_1{\mathrm{\&alpha;}}_2+2{\mathrm{\&alpha;}}_1^3$

${\mathrm{\&gamma;}}_4={\mathrm{\&alpha;}}_4-3{\mathrm{\&alpha;}}_2^2-4{\mathrm{\&alpha;}}_1{\mathrm{\&alpha;}}_3+12{\mathrm{\&alpha;}}_1^2{\mathrm{\&alpha;}}_2-6{\mathrm{\&alpha;}}_1^4$

${\mathrm{\&gamma;}}_5={\mathrm{\&alpha;}}_5-5{\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_1-10{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_2+20{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_1^2+30{\mathrm{\&alpha;}}_2^2-60{\mathrm{\&alpha;}}_2{\mathrm{\&alpha;}}_1^3+24{\mathrm{\&alpha;}}_1^5$

${\mathrm{\&gamma;}}_6={\mathrm{\&alpha;}}_6-6{\mathrm{\&alpha;}}_5{\mathrm{\&alpha;}}_1-15{\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_2+30{\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_1^2-10{\mathrm{\&alpha;}}_3^2+120{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_2{\mathrm{\&alpha;}}_1-120{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_1^3+30{\mathrm{\&alpha;}}_2^3-$

$270{\mathrm{\&alpha;}}_2^2{\mathrm{\&alpha;}}_1^2+360\mathrm{\&alpha;}2{\mathrm{\&alpha;}}_1^4-120{\mathrm{\&alpha;}}_1^6$

Each rank cumulant Δ S of node injecting power in the present invention ^(k)by each node load power cumulant with the output power from photovoltaic cells cumulant addition obtains, wherein each node load power is by being normal distribution, and its single order cumulant equals its desired value, the variance that second order cumulant is normal distribution, and three to six rank cumulant are zero.

In the present invention, adopt formula by injecting at random each rank cumulant Δ S ^(k)obtain each rank cumulant Δ X of state variable ^(k), wherein representing matrix the matrix that k power of middle element forms.

The present invention's beneficial effect is compared with the prior art, the present invention proposes a kind of probability assessment method of distributed photovoltaic power access impact, take into full account the probability distribution formula feature of load power and distributed photovoltaic power output, can assess more accurately the impact of distributed power source access.

The present invention is applicable to the probability assessment of distributed photovoltaic power access impact.

Accompanying drawing explanation

Fig. 1 is the probability assessment method flow diagram of distributed photovoltaic power access impact.

Embodiment

The specific embodiment of the invention is as shown in Fig. 1 flow process.The present embodiment implementation step is as follows:

The first step, input initial data, comprise that the general trend such as line parameter circuit value, generator, load injecting power calculates required data, should provide in addition the data of related node injection rate random distribution, such as will provide its desired value and variance etc. to the load of normal distribution;

Second step, if there is photovoltaic generation in network, provide the observation intensity of illumination of a day 24 hours, these data can make the software of HOMER obtain by one, utilize this software by the light intensity value in each month, to produce light intensity value hourly in appointed place in advance;

The 3rd step, is knowing one day and can obtain after light intensity value hourly light intensity value mean value and the variance of 1 year every day, then by light intensity value and photovoltaic generation is meritorious exert oneself between relation, i.e. through type P between relation and reactive power and active power _mthe power output P of=rA η Computation distribution formula photo-voltaic power supply _minitial value, wherein A is the square formation gross area, η is the total photoelectric conversion efficiency of square formation, the intensity of illumination that r is certain area.

The 4th step, the trend that provides normal operation with certainty tidal current computing method distributes, thereby tries to achieve the state variable X on benchmark operating point ₀, Jacobian matrix J ₀, obtain sensitivity matrix S ₀;

The 5th step, each rank cumulant of Computation distribution formula photo-voltaic power supply power output stochastic variable.The power output P of distributed photovoltaic power _mprobability density function f (P _m) become Beta to distribute, in formula, α and β are the form parameters that Beta distributes, and Γ is Gamma function, R _mfor square formation peak power output.Adopt formula calculate the v rank square α of stochastic variable x _v, wherein f (x) is the probability density function of stochastic variable x.For simplifying, calculate, cumulant is only got front 6 rank, by each rank square, adopts following formula to ask for front 6 rank cumulant:

γ ₁＝α ₁＝m

${\mathrm{\&gamma;}}_2={\mathrm{\&alpha;}}_2-{\mathrm{\&alpha;}}_1^2$

${\mathrm{\&gamma;}}_3={\mathrm{\&alpha;}}_3-3{\mathrm{\&alpha;}}_1{\mathrm{\&alpha;}}_2+2{\mathrm{\&alpha;}}_1^3$

${\mathrm{\&gamma;}}_4={\mathrm{\&alpha;}}_4-3{\mathrm{\&alpha;}}_2^2-4{\mathrm{\&alpha;}}_1{\mathrm{\&alpha;}}_3+12{\mathrm{\&alpha;}}_1^2{\mathrm{\&alpha;}}_2-6{\mathrm{\&alpha;}}_1^4$

${\mathrm{\&gamma;}}_5={\mathrm{\&alpha;}}_5-5{\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_1-10{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_2+20{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_1^2+30{\mathrm{\&alpha;}}_2^2-60{\mathrm{\&alpha;}}_2{\mathrm{\&alpha;}}_1^3+24{\mathrm{\&alpha;}}_1^5$

${\mathrm{\&gamma;}}_6={\mathrm{\&alpha;}}_6-6{\mathrm{\&alpha;}}_5{\mathrm{\&alpha;}}_1-15{\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_2+30{\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_1^2-10{\mathrm{\&alpha;}}_3^2+120{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_2{\mathrm{\&alpha;}}_1-120{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_1^3+30{\mathrm{\&alpha;}}_2^3-$

$270{\mathrm{\&alpha;}}_2^2{\mathrm{\&alpha;}}_1^2+360\mathrm{\&alpha;}2{\mathrm{\&alpha;}}_1^4-120{\mathrm{\&alpha;}}_1^6$

The 6th step, according to each rank cumulant Δ S of the random injecting power of character computing node of cumulant ^(k).Each rank cumulant Δ S of node injecting power ^(k)by each node load power cumulant with the output power from photovoltaic cells cumulant addition obtains, wherein each node load power is by being normal distribution, and its single order cumulant equals its desired value, the variance that second order cumulant is normal distribution, and three to six rank cumulant are zero.

The 7th step, by injecting at random each rank cumulant Δ S ^(k)obtain each rank cumulant Δ X of state variable ^(k).Adopt formula by injecting at random each rank cumulant Δ S ^(k)obtain each rank cumulant Δ X of state variable ^(k), wherein representing matrix the matrix that k power of middle element forms.

The 8th step, by distribution function and the probability density function of Gram-Charlier progression computing mode variable Δ X.

The 9th step, calculates i days N probability P that node voltage is out-of-limit _k(k=1,2 ... N), ask for the probability P of this day total voltage out-of-limit _i:

The tenth step, repeats second to the 9th step, calculates the out-of-limit probability of system voltage every day in a year, and year dangerous hourage of expectation voltage can be obtained by following formula:

Claims (5)

1. a probability assessment method for distributed photovoltaic power access impact, is characterized in that, said method comprising the steps of:

1) type, the rated power initial data of input network line parameter, generator and load bus power and distributed power source;

2) the input distributed photovoltaic power access place observation intensity of illumination of a day 24 hours;

3) by formula P _mthe power output P of=rA η Computation distribution formula photo-voltaic power supply _minitial value, wherein A is the square formation gross area, η is the total photoelectric conversion efficiency of square formation, the intensity of illumination that r is certain area;

4) apply the trend distribution that traditional tidal current computing method calculates normal operation, obtain the state variable X on benchmark operating point ₀, Jacobian matrix J ₀;

5) each rank cumulant of Computation distribution formula photo-voltaic power supply power output stochastic variable;

6) according to each rank cumulant Δ S of the random injecting power of character computing node of cumulant ^(k);

7) by each rank cumulant Δ S of random injecting power ^(k)obtain each rank cumulant Δ X of state variable ^(k);

8) by distribution function and the probability density function of Gram-Charlier progression computing mode variable Δ X;

9) calculate i days N probability P that node voltage is out-of-limit _k(k=1,2 ... N), ask for the probability P of this day total voltage out-of-limit _i: $P_i=1-{\mathrm{\&Pi;}}_{k=1}^N(1-P_k);$

10) repeat second to the 9th step, calculate the out-of-limit probability of system voltage every day in a year, year dangerous hourage of expectation voltage can be obtained by following formula:

2. the probability assessment method of a kind of distributed photovoltaic power access impact according to claim 1, is characterized in that: the power output P of described distributed photovoltaic power _mprobability density function f (P _m) become Beta to distribute, in formula, α and β are the form parameters that Beta distributes, and Γ is Gamma function, R _mfor square formation peak power output.

3. the probability assessment method of a kind of distributed photovoltaic power access impact according to claim 1, is characterized in that: described method adopts formula calculate the v rank square α of stochastic variable x _v, wherein f (x) is the probability density function of stochastic variable x; For simplifying, calculate, cumulant is only got front 6 rank, by each rank square, adopts following formula to ask for front 6 rank cumulant;

γ ₁＝α ₁

${\mathrm{\&gamma;}}_2={\mathrm{\&alpha;}}_2-{\mathrm{\&alpha;}}_1^2$

${\mathrm{\&gamma;}}_3={\mathrm{\&alpha;}}_3-{3\mathrm{\&alpha;}}_1{\mathrm{\&alpha;}}_2+{2\mathrm{\&alpha;}}_1^3$

${\mathrm{\&gamma;}}_4={\mathrm{\&alpha;}}_4-{3\mathrm{\&alpha;}}_2^2-{4\mathrm{\&alpha;}}_1{\mathrm{\&alpha;}}_3+{12\mathrm{\&alpha;}}_1^2{\mathrm{\&alpha;}}_2-{6\mathrm{\&alpha;}}_1^4$

${\mathrm{\&gamma;}}_5={\mathrm{\&alpha;}}_5-{5\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_1-10{\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_2{+20\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_1^2+{30\mathrm{\&alpha;}}_2^2-{60\mathrm{\&alpha;}}_2{\mathrm{\&alpha;}}_1^3+{24\mathrm{\&alpha;}}_1^5$

$\begin{array}{c}{\mathrm{\&gamma;}}_6={\mathrm{\&alpha;}}_6-{6\mathrm{\&alpha;}}_5{\mathrm{\&alpha;}}_1-{15\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_2+{30\mathrm{\&alpha;}}_4{\mathrm{\&alpha;}}_1^2-{10\mathrm{\&alpha;}}_3^2+{120\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_2{\mathrm{\&alpha;}}_1-{120\mathrm{\&alpha;}}_3{\mathrm{\&alpha;}}_1^3+{30\mathrm{\&alpha;}}_2^3-\\ {270\mathrm{\&alpha;}}_2^2{\mathrm{\&alpha;}}_1^2+360\mathrm{\&alpha;}{2\mathrm{\&alpha;}}_1^4-120{\mathrm{\&alpha;}}_1^6\end{array}.$

4. the probability assessment method of a kind of distributed photovoltaic power access impact according to claim 1, is characterized in that: each rank cumulant Δ S of the random injecting power of described node ^(k)by each node load power cumulant with the output power from photovoltaic cells cumulant addition obtains, wherein each node load power is by being normal distribution, and its single order cumulant equals its desired value, the variance that second order cumulant is normal distribution, and three to six rank cumulant are zero.

5. the probability assessment method of a kind of distributed photovoltaic power access impact according to claim 1, is characterized in that: described method step 7) by each rank cumulant Δ S of random injecting power ^(k)obtain each rank cumulant of state variable, adopt formula wherein representing matrix the matrix that k power of middle element forms.