CN102709908B - Loss prediction method for large-scale wind power-accessed power grid - Google Patents

Abstract

The invention discloses a loss prediction method for a large-scale wind power-accessed power grid and belongs to the technical field of control and prediction of grid-connected power generation of a wind farm. The method comprises the following steps: determining an output probability distribution curve of the wind farm, as well as an output probability distribution curve and a load probability distribution curve of a generator according to the system data and wind farm data; calculating the expectation value of power injection of each node, as well as semi invariant of each order of wind farm injection power, generator injection power and the load power of each node; counting out a coefficient matrix according to the injection power expectation value of each node; counting out the semi invariant of each order of injection power of each node according to the obtained semi invariant; counting out loss power of each node according to the coefficient matrix and the semi invariant of each order of injection power of each node, so as to obtain a probability density function and a cumulative distribution function of the loss; and calculating out the loss electricity quantity in set time section according to the cumulative distribution function. Due to the adoption of the method, the loss of the large-scale wind power-accessed power grid can be accurately and quickly predicted.

Description

Network loss prediction method after large-scale wind power is connected into power grid

Technical Field

The invention belongs to the technical field of grid-connected power generation control and prediction of a wind power plant, and particularly relates to a method for predicting loss of a large-scale wind power after the large-scale wind power is connected into a power grid.

Background

With the vigorous development of new energy in the world and the development of scientific technology in a new and different day by day, the development of the wind, light and electricity industry in China is rapid, and the development of the wind, light and electricity grid-connected technology is closely related to the development of national economy. The research on the influence of large-scale wind and photovoltaic access on the power grid is of great significance. Under the power market mode, especially after large-scale wind-photovoltaic access, the mode and the mode of the power grid operation are changed greatly, a power grid dispatcher needs to adjust the power grid operation mode in time according to market requirements and wind-photovoltaic power generation characteristics, and the power grid loss is predicted simply, accurately and rapidly when economic load distribution is carried out, which is very important.

The traditional method for calculating and predicting the network loss in the power system mainly comprises two types: the method comprises the following steps of firstly, an approximate algorithm, wherein the method for solving the electric energy loss according to the maximum load loss time and the method for solving the electric energy loss by utilizing loss factors are the simplest method, and in addition, the method also comprises a root mean square current method, an equivalent current distribution method and the like; the other is an accurate algorithm which mainly comprises an equivalent node power method, a loss power accumulation method, a dynamic power flow method, a node voltage interpolation/fitting method and a loss power interpolation/fitting method similar to the dynamic power flow method.

The methods cannot effectively account for the influence of node injection power fluctuation on network loss, cannot reflect the change rule of network loss power in an actual system after large-scale wind photovoltaic access, and have the disadvantages of large calculation amount, difficult data preparation and difficulty in meeting the requirement of network loss analysis when the network loss is predicted. Obviously, the method is not suitable for predicting the network loss after large-scale wind-solar-electricity access. Therefore, it is particularly important to develop a network loss prediction method for large-scale wind power access.

Disclosure of Invention

The invention aims to provide a network loss prediction method after large-scale wind power is accessed into a power grid, which is used for solving the problem that the network loss prediction precision is not high due to the fact that the influence of node injection power fluctuation on network loss cannot be effectively considered in the conventional network loss prediction method after the wind power is accessed into the power grid.

In order to achieve the purpose, the technical scheme provided by the invention is that the method for predicting the network loss after the large-scale wind power is connected into the power grid is characterized by comprising the following steps:

step 1: inputting system data and wind farm data, wherein the system data and the wind farm data comprise rated power of a wind driven generator, rated wind speed of the wind driven generator, wind speed, cut-in wind speed, cut-out wind speed, wind speed distribution parameters, active output of a generator set, probability of normal operation of the generator set, an active output value of the generator set in normal operation, an expected value of load active power, variance of the load active power and a load active power value; the wind speed distribution parameters comprise shape parameters, scale parameters and position parameters;

step 2: determining a probability distribution curve of the output of the wind power plant, a probability distribution curve of the output of the generator and a probability distribution curve of the load;

the wind power plant output probability distribution curve is

A probability density function of $f\left(P_W\right)=\exp [-{\left(\frac{P_W-k_1{v}_0-k_2}{k_{1}c}\right)}^k]\frac{k}{k_{1}c}{\left(\frac{P_W-k_1{v}_0-k_2}{k_{1}c}\right)}^{k-1};$ Wherein, P_WFor outputting power to the wind power generator <math> <mrow> <msub> <mi>P</mi> <mi>W</mi> </msub> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>ci</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>k</mi> <mn>1</mn> </msub> <mi>v</mi> <mo>+</mo> <msub> <mi>k</mi> <mn>2</mn> </msub> <mo>,</mo> </mtd> <mtd> <msub> <mi>v</mi> <mi>ci</mi> </msub> <mo>&lt;</mo> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>r</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>P</mi> <mi>r</mi> </msub> <mo>,</mo> </mtd> <mtd> <msub> <mi>v</mi> <mi>r</mi> </msub> <mo>&lt;</mo> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>co</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <mi>v</mi> <mo>></mo> <msub> <mi>v</mi> <mi>co</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math> v is the wind speed, P_rIs rated power of wind power generator, v_rRated wind speed, v, of a wind turbine_ciFor cutting into the wind speed, v_coTo cut out the wind speed, k₁=P_r/(v_r-v_ci)，k₂=-k₁v_ciAnd f (v) is a probability density function of wind speed and

k is a shape parameter, c is a scale parameter, v₀Is a position parameter;

the probability distribution curve of the output of the generator is $P(P_G=x_i)=\left\{\begin{array}{cc}{p}_p,& x_i=C_p\\ 1-p_p& x_i=0\end{array}\right.;$ Wherein, P_GIs the active output of the generator set, p_pProbability of normal operation of the generator set, C_pThe active output value is the active output value when the generator set normally operates;

the probability distribution curve of the load is

Having a probability density function of

Wherein, mu_PFor the desired value of the active power of the load, delta_PIs the variance of the active power of the load, P_LIs a load active power value;

and step 3: calculating to obtain a probability distribution curve of the injection power of each node according to the probability distribution curve of the output of the wind power plant, the probability distribution curve of the output of the generator and the probability distribution curve of the load, integrating and averaging the probability distribution curve of each node in a [0, T ] time period to obtain the average value of the injection power of each node in the [0, T ] time period, namely the expected value of the injection power of each node, wherein T is a set value;

and 4, step 4: according to the expected value of the injection power of each node, using a formula

Find the coefficient matrix J₀I.e. loss power P of the network_LOSSRespectively calculating the injection power P of each node, and substituting the expected value of the injection power of each node into a formula after calculation to obtain a coefficient matrix J₀(ii) a Wherein,

for the net loss power in tidal current conditions,

P=[P₁,P₂,…,P_n]，P_i(i =1, 2.. n.) injects power for each node,

injecting expected values of power for each node, wherein n is the number of the nodes;

and 5: respectively calculating each-order semi-invariant of wind power plant injection power, generator injection power and load power of each node according to the probability distribution curve of wind power plant output, the probability distribution curve of generator output and the probability distribution curve of load;

step 6: according to the formula

Calculating each-order semi-invariant of the injection power delta P of each node; wherein, Δ P^(k)For the k-th order semi-invariant of the injection power deltap of each node,

injecting a k-th order semi-invariant of power for the wind farm at each node,

a k-th order semi-invariant of the generator injected power for each node,

is each order semi-invariant of the load power of each node; k is the order of the semi-invariant;

and 7: using formulas

Calculating the power loss increment delta P of each node_LOSSAnd using a formula

Calculating the network loss power P of each node_LOSS(ii) a Wherein,

is a coefficient matrix J₀The k-th power of the elements in (a),

the network loss power under the tidal current condition;

and 8: method for solving power loss P by utilizing Gram-Charlie series expansion_LOSSA probability density function and a cumulative distribution function of;

and step 9: using formulas

Find [0, T]Network loss W in time interval_LOSS；

P_LOSS(P) is the network loss power P_LOSSThe cumulative distribution function of (a).

The method establishes the probability models of the output of the wind power plant, the output of the generator and the load based on the probability theory method, describes the change rule of the loss power after the large-scale wind power is accessed, can quickly and accurately predict the loss, and solves the problem that the traditional loss prediction method cannot effectively predict the loss after the large-scale wind power is accessed.

Drawings

FIG. 1 is a flow chart of a method for predicting loss of a large-scale wind power grid after the wind power grid is connected;

FIG. 2 is an IEEE30 node system wiring diagram;

fig. 3 is a network loss probability distribution diagram.

Detailed Description

The preferred embodiments will be described in detail below with reference to the accompanying drawings. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.

Example 1

Fig. 1 is a flow chart of a network loss prediction method after large-scale wind power is connected to a power grid. In fig. 1, the method for predicting the loss of the large-scale wind power after the large-scale wind power is connected to the power grid, provided by the invention, comprises the following steps:

step 1: inputting system data and wind farm data, wherein the system data and the wind farm data comprise rated power of a wind driven generator, rated wind speed of the wind driven generator, wind speed, cut-in wind speed, cut-out wind speed, wind speed distribution parameters, active output of a generator set, probability of normal operation of the generator set, an active output value of the generator set in normal operation, an expected value of load active power, variance of the load active power and a load active power value; the wind speed distribution parameters include shape parameters, scale parameters, and location parameters.

Step 2: and determining a probability distribution curve of the output of the wind power plant, a probability distribution curve of the output of the generator and a probability distribution curve of the load.

(1) And determining a probability distribution curve of the output of the wind power plant.

1) The probability distribution of the wind speed is calculated. When the wind speed conforms to the Weibull distribution, the probability density function of its wind speed can be described as:

wherein v is the wind speed; k. c and v₀Is 3 parameters of Weibull distribution, k is a shape parameter which reflects the characteristics of wind speed distribution, c is a scale parameter which reflects the size of the average wind speed in the area, v₀Is a location parameter.

2) And establishing a functional relation between the output power of the wind driven generator and the wind speed. The expression of the output power of the wind driven generator is as follows: <math> <mrow> <msub> <mi>P</mi> <mi>W</mi> </msub> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>ci</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>k</mi> <mn>1</mn> </msub> <mi>v</mi> <mo>+</mo> <msub> <mi>k</mi> <mn>2</mn> </msub> <mo>,</mo> </mtd> <mtd> <msub> <mi>v</mi> <mi>ci</mi> </msub> <mo>&lt;</mo> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>r</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>P</mi> <mi>r</mi> </msub> <mo>,</mo> </mtd> <mtd> <msub> <mi>v</mi> <mi>r</mi> </msub> <mo>&lt;</mo> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>co</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <mi>v</mi> <mo>></mo> <msub> <mi>v</mi> <mi>co</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math> wherein k is₁=P_r/(v_r-v_ci)；k₂=-k₁v_ci；P_rThe rated power of the wind driven generator; v. of_rRated wind speed; v. of_ciTo cut into the wind speed; v. of_coTo cut out the wind speed.

3) When the wind speed and the active power output by the wind driven generator satisfy a linear relation, the probability distribution curve of the active power output of the wind driven generator can be obtained according to the two formulas as follows:

<math> <mrow> <mi>F</mi> <mrow> <mo>(</mo> <msub> <mi>P</mi> <mi>W</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <msubsup> <mo>&Integral;</mo> <msub> <mi>v</mi> <mn>0</mn> </msub> <msub> <mi>v</mi> <mi>ci</mi> </msub> </msubsup> <mi>f</mi> <mrow> <mo>(</mo> <mi>v</mi> <mo>)</mo> </mrow> <mi>dv</mi> <mo>+</mo> <msubsup> <mo>&Integral;</mo> <msub> <mi>v</mi> <mi>ci</mi> </msub> <mfrac> <mrow> <msub> <mi>P</mi> <mi>W</mi> </msub> <mo>-</mo> <msub> <mi>k</mi> <mn>2</mn> </msub> </mrow> <msub> <mi>k</mi> <mn>1</mn> </msub> </mfrac> </msubsup> <mi>f</mi> <mrow> <mo>(</mo> <mi>v</mi> <mo>)</mo> </mrow> <mi>dv</mi> <mo>,</mo> </mrow> </math>

the probability density function is then:

$f\left(P_W\right)=\exp [-{\left(\frac{P_W-k_1{v}_0-k_2}{k_{1}c}\right)}^k]\frac{k}{k_{1}c}{\left(\frac{P_w-k_1{v}_0-k_2}{k_{1}c}\right)}^{k-1}.$

(2) and calculating a probability distribution curve of the output of the generator.

The generator output is described by adopting a two-state generator set model, namely a (0-1) model of a discrete variable, namely the generator set only has two states of normal operation and forced failure of the fault, so that the probability distribution curve of the generator output is $P(P_G=x_i)=\left\{\begin{array}{cc}{p}_p,& x_i=C_p\\ 1-p_p& x_i=0\end{array}\right.;$ Wherein, P_GIs the active output of the generator set, p_pIs a generator setProbability of constant operation, C_pThe active output value is the active output value when the generator set normally operates.

(3) A probability distribution curve of the load is calculated.

The load is assumed to follow a normal random variable distribution. The expected value and variance value of the active power of the load are respectively mu_PAnd delta_PThe probability distribution curve of the load is

Having a probability density function of <math> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <msub> <mi>P</mi> <mi>L</mi> </msub> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <msqrt> <mn>2</mn> <mi>&pi;</mi> <msub> <mi>&delta;</mi> <mi>P</mi> </msub> </msqrt> </mfrac> <mi>exp</mi> <mo>[</mo> <mo>-</mo> <mfrac> <msup> <mrow> <mo>(</mo> <msub> <mi>P</mi> <mi>L</mi> </msub> <mo>-</mo> <msub> <mi>&mu;</mi> <mi>P</mi> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mrow> <mn>2</mn> <msup> <msub> <mi>&delta;</mi> <mi>P</mi> </msub> <mn>2</mn> </msup> </mrow> </mfrac> <mo>]</mo> <mo>,</mo> </mrow> </math> P_LIs the load active power value.

And step 3: calculating to obtain a probability distribution curve of the injection power of each node according to the probability distribution curve of the output of the wind power plant, the probability distribution curve of the output of the generator and the probability distribution curve of the load, integrating and averaging the probability distribution curve of each node in the period of [0, T ], and obtaining the average value of the injection power of each node in the period of [0, T ], namely the expected value of the injection power of each node, wherein T is a set value.

The node injection power is equal to the sum of the wind power plant output and the generator output at the node, and then the load value at the node is subtracted. Therefore, according to the obtained probability distribution curves of the output of the wind power plant, the output of the generator and the load, the probability distribution curve of the injection power of each node can be obtained, the probability distribution curves of each node are integrated and averaged in the period of [0, T ], and then the mean value of the injection power of each node in the period of [0, T ], namely the expected value of the injection power of each node, can be obtained.

The expected value of the injected power of the node is actually that the injected power of each node is taken to be [0, T ] of the node]Mean value of time interval injection power

The power flow is called as mean power flow, and the network loss power under the condition of the power flow is recorded as

For a continuous random variable X, the probability density function is f (X), if integratedConvergence, then called the integral value

Is the mathematical expectation of X. For a discrete random variable X, its probability distribution is P (X = X)_i)=p_iIf, if

Then call

Is the mathematical expectation of X.

And 4, step 4: according to the expected value of the injection power of each node, using a formula

System of solutionNumber matrix J₀I.e. loss power P of the network_LOSSRespectively calculating the injection power P of each node, and substituting the expected value of the injection power of each node into a formula after calculation to obtain a coefficient matrix J₀。

If the fluctuation range of the injection power of each node is not very large, the network loss power in any operation mode can be represented as follows:the formula is a linear function expression between the network loss power and the node injection power. Wherein,

for net loss power under tidal current conditions, J₀Δ P is the power loss increase of the network, using Δ P_LOSSIs expressed as Δ P_LOSS=J₀ΔP。

P=[P₁,P₂,…,P_n]，P_i(i =1, 2.. n.) injects power for each node,

the expected value of power is injected for each node, and n is the number of nodes.

And 5: and respectively calculating each-order semi-invariant of wind power plant injection power, generator injection power and load power of each node according to the probability distribution curve of wind power plant output, the probability distribution curve of generator output and the probability distribution curve of load.

Calculating semi-invariants of each order through a probability distribution curve is the prior art in the probability method, so that the semi-invariants of each order of wind power plant injection power, generator injection power and load power of each node are calculated respectively according to the probability distribution curve of wind power plant output, the probability distribution curve of generator output and the probability distribution curve of load, and the method is a common technology for technicians in the field and is not described in the embodiment.

Step 6: according to the formula

Calculating each-order semi-invariant of the injection power delta P of each node; wherein, Δ P^(k)For the k-th order semi-invariant of the injection power deltap of each node,

injecting a k-th order semi-invariant of power for the wind farm at each node,

a k-th order semi-invariant of the generator injected power for each node,

is each order semi-invariant of the load power of each node; k is the order of the semi-invariant.

The random variable vector of the injection power of each node is determined by the following formula:

wherein, Δ P_W，ΔP_G，ΔP_LRespectively a random variable vector (existing at a wind power plant node) of the output power of the wind power plant, a random variable vector of the output power of the generator of each node and a random variable vector, symbol of the load power

Representing a convolution operation, which may be implemented with a semi-invariant herein. Thus, depending on the nature of the semi-invariant, each order of power is loaded by a node with a semi-invariant

And each step of generator power

Each-order semi-invariant delta P of node injection power can be obtained^(k)For windThe grid-connected point of the electric field needs to be added with each-order semi-invariant of the output of the wind power plant, namely: <math> <mrow> <mi>&Delta;</mi> <msup> <mi>P</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </msup> <mo>=</mo> <mi>&Delta;</mi> <msubsup> <mi>P</mi> <mi>W</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </msubsup> <mo>+</mo> <mi>&Delta;</mi> <msubsup> <mi>P</mi> <mi>G</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </msubsup> <mo>+</mo> <mi>&Delta;</mi> <msubsup> <mi>P</mi> <mi>L</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </msubsup> <mo>.</mo> </mrow> </math>

and 7: using formulas

Calculating the power loss increment delta P of each node_LOSSAnd using a formula

Calculating the network loss power P of each node_LOSS(ii) a Wherein,

is a matrix J₀The k-th power of the elements in (a),

the network loss power under the tidal current condition.

And 8: method for solving power loss P by utilizing Gram-Charlie series expansion_LOSSA probability density function and a cumulative distribution function.

The Gram-Charlier series expansion is often used in the random production simulation of power systemsThe distribution function of the random variable is expressed as a series consisting of derivatives of the orders of the normal random variable, and the coefficient of the series consists of semi-invariants of the orders of the random variable. The principle is as follows: for any random variable X, assuming that the expected value is mu and the standard deviation is delta, the normalized random variable is

Comprises the following steps:

<math> <mrow> <mover> <mi>X</mi> <mo>&OverBar;</mo> </mover> <mo>=</mo> <mrow> <mo>(</mo> <mi>X</mi> <mo>-</mo> <mi>&mu;</mi> <mo>)</mo> </mrow> <mo>/</mo> <mi>&delta;</mi> </mrow> </math>

let f (x) be a normalized random variable

According to Gram-Charlier series expansion theory, f (x) has the form:

f(x)=φ(x)+(c₁/1!)φ′(x)+(c₂/2!)φ″(x)+(c₃/3!)φ″′(x)+…

where φ (x) is a probability density function of a standard normal distribution. Coefficient c_kComprises the following steps:

c₁=0

c₂=0

<math> <mrow> <msub> <mi>c</mi> <mn>3</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <msub> <mi>&beta;</mi> <mn>3</mn> </msub> <msup> <mi>&delta;</mi> <mn>3</mn> </msup> </mfrac> </mrow> </math>

<math> <mrow> <msub> <mi>c</mi> <mn>4</mn> </msub> <mo>=</mo> <mfrac> <msub> <mi>&beta;</mi> <mn>4</mn> </msub> <msup> <mi>&delta;</mi> <mn>4</mn> </msup> </mfrac> <mo>-</mo> <mn>3</mn> </mrow> </math>

<math> <mrow> <msub> <mi>c</mi> <mn>5</mn> </msub> <mo>=</mo> <mo>-</mo> <mfrac> <msub> <mi>&beta;</mi> <mn>5</mn> </msub> <msup> <mi>&delta;</mi> <mn>5</mn> </msup> </mfrac> <mo>+</mo> <mn>10</mn> <mfrac> <msub> <mi>&beta;</mi> <mn>3</mn> </msub> <msup> <mi>&delta;</mi> <mn>3</mn> </msup> </mfrac> </mrow> </math>

<math> <mrow> <msub> <mi>c</mi> <mn>6</mn> </msub> <mo>=</mo> <mfrac> <msub> <mi>&beta;</mi> <mn>6</mn> </msub> <msup> <mi>&delta;</mi> <mn>6</mn> </msup> </mfrac> <mo>-</mo> <mn>15</mn> <mfrac> <msub> <mi>&beta;</mi> <mn>4</mn> </msub> <msup> <mi>&delta;</mi> <mn>4</mn> </msup> </mfrac> <mo>+</mo> <mn>30</mn> </mrow> </math>

in the formula, beta_v(v =1.2.…) is the center distance of each order of a random variable, which can be expressed as a polynomial of a semi-invariant; δ is the standard deviation of the random variable.

Using Gram-Charlie series expansion, through one calculationCan obtain the desired P_LOSSThen obtaining P by integrating the probability density function_LOSSThe cumulative distribution function of (a). Compared with the traditional Monte Carlo method, the method for solving the P by using Gram-Charlie series expansion_LOSSThe probability density function and the cumulative distribution function of (2) can shorten the calculation time.

And step 9: using formulas

Find [0, T]Network loss W in time interval_LOSS。

Find P_LOSSAfter the probability distribution function and the cumulative distribution function, the formula is adoptedCan find out [0, T]Network loss W in time interval_LOSS. P is the probability, P_LOSS(P) is the network loss power P_LOSSThe cumulative distribution function of (a).

Example 2

According to the above method, the present embodiment adopts the IEEE30 node system (fig. 2 is the system wiring diagram) as shown in fig. 2 as a verification model, and the analysis is as follows:

suppose a wind farm is connected to the 29 th node of an IEEE30 power saving system through a transformer and a 110kV line, and the line parameter of the wind farm connected to the system is 12.6+ j24.96 omega. The mathematical expectation of the injected power at each node of the system can be derived from the probability distribution curves of the wind farm contribution, the generator contribution and the load, as shown in table 1.

TABLE 1 mathematical expectation of injected power at each node

The variation of the network loss of the system in one day [0,24h ] (h represents hour) can be known by calculating the semi-invariant of the network loss power and using Gram-Charlier series expansion to obtain the probability distribution curve of the network loss of the system in one day [0,24h ] (h represents hour) (fig. 3), as shown in table 2.

TABLE 2 network loss variation for system [0,24h ]

The above example analysis shows that: the method solves the problems that the traditional method is large in calculated amount and difficult in data preparation, cannot reflect the influence of large-scale wind-photovoltaic access on network loss, cannot meet the requirement of network loss analysis and the like, establishes a probability model of wind power plant output, generator output and load by using a probability theory correlation method, deduces the mathematical relationship between network loss power and node injection power by using a linearization method, and obtains the probability distribution of the network loss power by adopting a method combining a semi-invariant and Gram-Charlier series expansion, so that the network loss electric quantity can be accurately and conveniently predicted on the basis, and the calculation time can be effectively reduced.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (1)

1. A method for predicting the loss of a large-scale wind power grid is characterized by comprising the following steps:

step 1: inputting system data and wind farm data, wherein the system data and the wind farm data comprise rated power of a wind driven generator, rated wind speed of the wind driven generator, wind speed, cut-in wind speed, cut-out wind speed, wind speed distribution parameters, active output of a generator set, probability of normal operation of the generator set, an active output value of the generator set in normal operation, an expected value of load active power, variance of the load active power and a load active power value; the wind speed distribution parameters comprise shape parameters, scale parameters and position parameters;

step 2: determining a probability distribution curve of the output of the wind power plant, a probability distribution curve of the output of the generator and a probability distribution curve of the load;

the wind power plant output probability distribution curve is

A probability density function of $f\left(P_W\right)=\exp [-{\left(\frac{P_W-k_1{v}_0-k_2}{k_{1}c}\right)}^k]\frac{k}{k_{1}c}{\left(\frac{P_W-k_1{v}_0-k_2}{k_{1}c}\right)}^{k-1};$ Wherein, P_WFor outputting power to the wind power generator <math> <mrow> <msub> <mi>P</mi> <mi>W</mi> </msub> <mo>=</mo> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>ci</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>k</mi> <mn>1</mn> </msub> <mi>v</mi> <mo>+</mo> <msub> <mi>k</mi> <mn>2</mn> </msub> <mo>,</mo> </mtd> <mtd> <msub> <mi>v</mi> <mi>ci</mi> </msub> <mo>&lt;</mo> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>r</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>P</mi> <mi>r</mi> </msub> <mo>,</mo> </mtd> <mtd> <msub> <mi>v</mi> <mi>r</mi> </msub> <mo>&lt;</mo> <mi>v</mi> <mo>&le;</mo> <msub> <mi>v</mi> <mi>co</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> <mo>,</mo> </mtd> <mtd> <mi>v</mi> <mo>></mo> <msub> <mi>v</mi> <mi>co</mi> </msub> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math> v is the wind speed, P_rIs rated power of wind power generator, v_rRated wind speed, v, of a wind turbine_ciFor cutting into the wind speed, v_coTo cut out the wind speed, k₁=P_r/(v_r-v_ci)，k₂=-k₁v_ciAnd f (v) is a probability density function of wind speed and

k is a shape parameter, c is a scale parameter, v₀Is a position parameter;

the probability distribution curve of the output of the generator is $P(P_G=x_i)=\left\{\begin{array}{cc}{p}_p,& x_i=C_p\\ 1-p_p& x_i=0\end{array}\right.;$ Wherein, P_GIs the active output of the generator set, p_pProbability of normal operation of the generator set, C_pThe active output value is the active output value when the generator set normally operates;

the probability distribution curve of the load is

Having a probability density function of

Wherein, mu_PFor the desired value of the active power of the load, delta_PIs the variance of the active power of the load, P_LIs a load active power value;

and step 3: calculating to obtain a probability distribution curve of the injection power of each node according to the probability distribution curve of the output of the wind power plant, the probability distribution curve of the output of the generator and the probability distribution curve of the load, integrating and averaging the probability distribution curve of each node in a [0, T ] time period to obtain the average value of the injection power of each node in the [0, T ] time period, namely the expected value of the injection power of each node, wherein T is a set value;

and 4, step 4: according to the expected value of the injection power of each node, using a formula

Find the coefficient matrix J₀I.e. loss power P of the network_LOSSRespectively calculating the injection power P of each node, and substituting the expected value of the injection power of each node into a formula after calculation to obtain a coefficient matrix J₀(ii) a Wherein, <math> <mrow> <msub> <mi>P</mi> <mi>LOSS</mi> </msub> <mo>=</mo> <msubsup> <mi>P</mi> <mi>LOSS</mi> <mn>0</mn> </msubsup> <mo>+</mo> <msub> <mi>J</mi> <mn>0</mn> </msub> <mi>&Delta;P</mi> <mo>,</mo> </mrow> </math>

for the net loss power in tidal current conditions, <math> <mrow> <mi>&Delta;P</mi> <mo>=</mo> <mo>[</mo> <msub> <mi>P</mi> <mn>1</mn> </msub> <mo>-</mo> <msubsup> <mi>P</mi> <mn>1</mn> <mn>0</mn> </msubsup> <mo>,</mo> <msub> <mi>P</mi> <mn>2</mn> </msub> <mo>-</mo> <msubsup> <mi>P</mi> <mn>2</mn> <mn>0</mn> </msubsup> <mo>,</mo> <mo>&CenterDot;</mo> <mo>&CenterDot;</mo> <mo>&CenterDot;</mo> <mo>,</mo> <msub> <mi>P</mi> <mi>n</mi> </msub> <mo>-</mo> <msubsup> <mi>P</mi> <mi>n</mi> <mn>0</mn> </msubsup> <mo>]</mo> <mo>,</mo> </mrow> </math> P=[P₁,P₂,…,P_n]，P_ithe power is injected for each of the nodes,

injecting a desired value of power for each node, i =1, 2.. and n is the number of nodes;

and 5: respectively calculating each-order semi-invariant of wind power plant injection power, generator injection power and load power of each node according to the probability distribution curve of wind power plant output, the probability distribution curve of generator output and the probability distribution curve of load;

step 6: according to the formula

Calculating each-order semi-invariant of the injection power delta P of each node; wherein, Δ P^(k)For the k-th order semi-invariant of the injection power deltap of each node,

injecting a k-th order semi-invariant of power for the wind farm at each node,a k-th order semi-invariant of the generator injected power for each node,

is each order semi-invariant of the load power of each node; k is the order of the semi-invariant;

and 7: using formulas

Calculating the power loss increment delta P of each node_LOSSAnd using a formulaCalculating the network loss power P of each node_LOSS(ii) a Wherein,

is a coefficient matrix J₀The k-th power of the elements in (a),

the network loss power under the tidal current condition;

and 8: method for solving power loss P by utilizing Gram-Charlie series expansion_LOSSA probability density function and a cumulative distribution function of;

and step 9: using formulas

Find [0, T]Network loss W in time interval_LOSS；P_LOSS(P) is the network loss power P_LOSSThe cumulative distribution function of (a).

Images