CN105373858A - Wind power plant active power optimization method based on wind speed time sequence decomposition - Google Patents

Abstract

The invention discloses a wind power plant active power optimization method based on wind speed time sequence decomposition. The method comprises the following steps that step one: a wind speed prediction curve of the next control cycle of a wind power plant is converted into a discrete wind speed sequence; step two: wind speed values corresponding to all elements of the discrete wind speed sequence are regarded as constant natural wind speed, and the active power reference value and initial time thereof of each machine set are calculated respectively under the effect of constant natural wind speed; and step three: an active power reference value control curve of each machine set under the effect of the wind speed sequence is generated by the active power reference value and the initial time thereof of each machine set, and each machine set operates according to the control curve. According to the optimization method, active power optimization of the wind power plant under dynamic wind speed is realized by using lower amount of calculation and less hardware resources so that a scheme with practical value is formed, generation capacity of the wind power plant can be enhanced and the generation benefits of the wind power plant can be enhanced.

Description

A kind of active power of wind power field optimization method decomposed based on wind speed sequential

Technical field

The present invention relates to wind energy turbine set control technology field, particularly a kind of active power of wind power field optimization method decomposed based on wind speed sequential.

Background technology

In recent years, the Wind Power Generation Industry of China remains strong growth, along with the development of wind-powered electricity generation, has the land resources of good wind energy fewer and feweri, and further develop for supporting Wind Power Generation Industry, extensive development marine wind electric field becomes inexorable trend.Current marine wind electric field construction cost is higher, improves the generating efficiency of marine wind electric field, increases the economic benefit of wind-powered electricity generation owner, is one of key issue of marine wind electric field operation.

Compared with landwid electric field, marine wind electric field is very violent by wake effect, traditional separate unit wind turbine generator maximum wind energy capture, larger wind energy is absorbed by causing upwind unit, lower wind direction unit input wind speed reduces, thus loss active power of wind power field, research shows, traditional unit maximal wind-energy capture scheme, the marine wind electric field active power loss caused due to wake effect can up to 30%, although by wind energy turbine set microcosmic structure, Wind turbines layout optimization effectively can reduce the impact of wake effect, but due to factor restrictions such as wind energy turbine set place and construction costs, distance between Wind turbines is generally between 5 to 7 times of impeller diameter, wake effect is still obvious on the impact of active power of wind power field.

Both at home and abroad some scholars have carried out a series of research based on wake effect to active power of wind power field optimization, define multiple wake model, existing research substantially concentrate on wind speed constant time, wake effect is on the impact of active power of wind power field and ameliorative way aspect thereof.But when wind speed dynamic change, the research of the active power optimization aspect of wind energy turbine set is still immature, it is very strong non-linear that its main cause is that Wind turbines has, by traditional method of thinking, the wind speed of longer a period of time is integrally investigated, set up active power of wind power field optimization method, because may combining of time dimension is extremely many, make equation solution calculated amount large, corresponding control strategy realizes difficulty, is unfavorable for practical application.

Summary of the invention

In order to solve the problems of the technologies described above, the invention provides the active power of wind power field optimization method decomposed based on wind speed sequential that a kind of calculated amount is little, cost is low, practical value is high.

The technical scheme that the present invention solves the problem is: a kind of active power of wind power field optimization method decomposed based on wind speed sequential, comprises the following steps:

Step one: the discrete wind series forecasting wind speed curve of next for wind energy turbine set control cycle being converted to chronologically 30 elements;

Step 2: air speed value corresponding for each for discrete wind series element is considered as constant natural wind speed, based on wake effect, calculates under each constant natural wind speed effect, the active power reference value of each unit respectively; Meanwhile, based on wake flow propagation delay, calculate respectively under each constant natural wind speed effect, each unit active power reference value starts the moment acted on;

Step 3: by each unit active power reference value and effect moment, each unit active power reference value controlling curve under generating discrete wind series effect, each unit runs according to this controlling curve.

The above-mentioned active power of wind power field optimization method decomposed based on wind speed sequential, in described step one, the conversion formula of discrete wind series s (m) is:

$s\left(m\right)=\frac{1}{T_s}{\∫}_{t_N+(m-1)T_s}^{t_N+{\mathrm{mT}}_s}{s}_c\left(t\right)dt$ ①

Wherein, m is natural number, and its span is 1≤m≤31, s _ct () is t predicting wind speed of wind farm value, t _nrepresent the initial time of next control cycle, T represents the duration that wind energy turbine set control cycle is corresponding, T _swind speed discretize interval time, T _s=T/30.

The above-mentioned active power of wind power field optimization method decomposed based on wind speed sequential, in described step 2, the calculation procedure of the active power reference value of each unit is as follows:

1) with the axial inducible factor a of unit i _ifor variable, then the power coefficient of unit i and thrust coefficient for:

$C_p^i=4{(1-a_i)}^2{a}_i$ ②

$C_T^i=4(1-a_i)a_i$ ③

2) m in discrete wind series s (m) gets occurrence n, makes n=1;

3) establish wind energy turbine set to be subject to constant natural wind speed s (n) effect, use v _irepresent the input wind speed of unit i, if v ₁be upwind border unit, unit i+1 is unit i leeward upwards First unit, 4. 5. calculates the input wind speed of unit i+1 by formula with formula:

v ₁＝s(n)④

$v_{i+1}=v_i\·\[1-(1-\sqrt{1-C_T^i}){\left(\frac{r_i}{r_i+kx}\right)}^2\]$ ⑤

Wherein, k is wind energy turbine set terrain rough factor, r _ifor unit i impeller radius, x is the distance that unit i hub centre and unit i+1 hub centre project on wind direction;

4) each unit active power is calculated as follows:

⑥

Wherein, P _irepresent the active power of unit i, v _ratedrepresent unit wind rating, v _in, v _cutrepresent the incision of unit, cut-out wind speed; ρ represents atmospheric density, represent the rated power of unit i;

5) active power of wind power field Optimized model is set up as follows:

$\begin{array}{c}{P}_{all}=\underset{a_i}{\max }\left(\underset{i-1}{\overset{N}{\Σ}}{P}_i\right)\\ s.t.\\ v_1=s\left(n\right)\\ v_{i+1}=v_i\·\[1-\left(1-\sqrt{1-C_T^i}\right){\left(\frac{r_i}{r_i+kx}\right)}^2\]\\ 0\≤a_i\≤\frac{1}{3}\\ 0.2P_{rated}^i\≤P_i\≤P_{rated}^i\end{array}$ ⑦

Wherein, P _allrepresent active power of wind power field, i.e. wind energy turbine set each unit active power sum, N represents unit sum in wind energy turbine set, represent by optimizing a _ithe formula in bracket is made to obtain maximal value;

6) solution procedure 5) given by equation, obtain the optimization solution of unit i inducible factor, and substituted into formula 2.-5., draw the concrete value of unit i power coefficient, thrust coefficient, input wind speed;

7) unit i power coefficient value is substituted into following formula, then when to obtain nature wind speed be s (n), the active power reference value of unit i

$P_{ref}^{i,n}=\frac{1}{2}{\mathrm{\ρ\πr}}_i^2{v}_i^3{C}_p^i$ ⑧

Make i value travel through wind energy turbine set each unit, then when to obtain nature wind speed be s (n), the active power reference value of each unit;

8) judge the value size of n, if n≤30, then make n=n+1, jump to step 3), otherwise terminate computing.

The above-mentioned active power of wind power field optimization method decomposed based on wind speed sequential, in described step 2, the computing method that the active power reference value of each unit starts the moment acted on are as follows: when the m in discrete wind series s (m) gets occurrence n, calculation of wind speed s (n) propagates into the time delay D of unit i from wind energy turbine set upwind border unit ^i,n, then when wind speed is s (n), the active power reference value of unit i the initiation moment for

$t_{ref}^{i,n}=t_N+D^{i,n}$ ⑨。

The above-mentioned active power of wind power field optimization method decomposed based on wind speed sequential, starting in the moment computing method acted on of the active power reference value of each unit, wind speed s (n) propagates into the time delay D of unit i from wind energy turbine set upwind border unit ^i,ncomputing method as follows:

1) wind presses wind direction propagation, on wind speed travel path, asks for the distance of wind energy turbine set upwind border unit impeller central and the subpoint of unit i impeller central on wind direction, is designated as L;

2) D is calculated as follows ^i,n, and make i value travel through each unit of wind energy turbine set, each natural number in n value traversal 1≤n≤31,

$D^{i,n}=\frac{L}{s\left(n\right)}$ ⑩。

The above-mentioned active power of wind power field optimization method decomposed based on wind speed sequential, in described step 3, the generating mode of unit active power reference value controlling curve is as follows: by i value, and n value is identical respectively with composition control amount its implication is in the moment, the active power reference value of unit i is to the i value determined, will value press the ascending arrangement of n, arrive between time, the active power reference value of unit i maintains constant, thus obtain the active power reference value controlling curve of unit i; Make i value travel through each unit of wind energy turbine set, each natural number in m value traversal 1≤m≤31, thus obtain wind energy turbine set each unit active power reference value controlling curve.

Beneficial effect of the present invention is: the prediction of wind speed Curve transform of next for wind energy turbine set control cycle is discrete wind series by the present invention, after air speed value corresponding for each for discrete wind series element is considered as constant natural wind speed, calculate each unit active power reference value under constant natural wind speed effect and start moment of acting on, and then each unit active power reference value controlling curve under the effect of generation wind series, each unit runs according to this controlling curve; Lower calculated amount, the less hardware resource of this optimization method, realizes active power of wind power field optimization under Dynamic Wind Speed, forms one and have practical value scheme, promote wind energy turbine set generated energy, improve wind energy turbine set gene-ration revenue.

Accompanying drawing explanation

Fig. 1 is general flow chart of the present invention.

Fig. 2 is the process flow diagram of step 3 of the present invention.

Embodiment

Below in conjunction with drawings and Examples, the present invention is further illustrated.

As shown in Figure 1, 2, the present invention includes following steps:

Step one: the discrete wind series forecasting wind speed curve of next for wind energy turbine set control cycle being converted to chronologically 30 elements.

The conversion formula of discrete wind series s (m) is:

$s\left(m\right)=\frac{1}{T_s}{\∫}_{t_N+(m-1)T_s}^{t_N+{\mathrm{mT}}_s}{s}_c\left(t\right)dt$ ①

Wherein, m is natural number, and its span is 1≤m≤31, s _ct () is t predicting wind speed of wind farm value, t _nrepresent the initial time of next control cycle, T represents the duration that wind energy turbine set control cycle is corresponding, T _swind speed discretize interval time, T _s=T/30.

Step 2: air speed value corresponding for each for discrete wind series element is considered as constant natural wind speed, based on wake effect, calculates under each constant natural wind speed effect, the active power reference value of each unit respectively; Meanwhile, based on wake flow propagation delay, calculate respectively under each constant natural wind speed effect, each unit active power reference value starts the moment acted on.

The calculation procedure of the active power reference value of each unit is as follows:

1) with the axial inducible factor a of unit i _ifor variable, then the power coefficient of unit i and thrust coefficient for:

$C_p^i=4{(1-a_i)}^2{a}_i$ ②

$C_T^i=4(1-a_i)a_i$ ③

2) m in discrete wind series s (m) gets occurrence n, makes n=1;

3) establish wind energy turbine set to be subject to constant natural wind speed s (n) effect, use v _irepresent the input wind speed of unit i, if v ₁be upwind border unit, unit i+1 is unit i leeward upwards First unit, 4. 5. calculates the input wind speed of unit i+1 by formula with formula:

v ₁＝s(n)④

$v_{i+1}=v_i\·\[1-(1-\sqrt{1-C_T^i}){\left(\frac{r_i}{r_i+kx}\right)}^2\]$ ⑤

Wherein, k is wind energy turbine set terrain rough factor, r _ifor unit i impeller radius, x is the distance that unit i hub centre and unit i+1 hub centre project on wind direction;

4) each unit active power is calculated as follows:

⑥

Wherein, P _irepresent the active power of unit i, v _ratedrepresent unit wind rating, v _in, v _cutrepresent the incision of unit, cut-out wind speed; ρ represents atmospheric density, represent the rated power of unit i;

5) active power of wind power field Optimized model is set up as follows:

$\begin{array}{c}{P}_{all}=\underset{a_i}{\max }\left(\underset{i-1}{\overset{N}{\Σ}}{P}_i\right)\\ s.t.\\ v_1=s\left(n\right)\\ v_{i+1}=v_i\·\[1-\left(1-\sqrt{1-C_T^i}\right){\left(\frac{r_i}{r_i+kx}\right)}^2\]\\ 0\≤a_i\≤\frac{1}{3}\\ 0.2P_{rated}^i\≤P_i\≤P_{rated}^i\end{array}$ ⑦

Wherein, P _allrepresent active power of wind power field, i.e. wind energy turbine set each unit active power sum, N represents unit sum in wind energy turbine set, represent by optimizing a _ivalue makes the formula in bracket obtain maximal value;

6) solution procedure 5) given by equation, obtain the optimization solution of unit i inducible factor, and substituted into formula 2.-5., draw the concrete value of unit i power coefficient, thrust coefficient, input wind speed;

7) unit i power coefficient value is substituted into following formula, then when to obtain nature wind speed be s (n), the active power reference value of unit i

$P_{ref}^{i,n}=\frac{1}{2}{\mathrm{\ρ\πr}}_i^2{v}_i^3{C}_p^i$ ⑧

Make i value travel through wind energy turbine set each unit, then when to obtain nature wind speed be s (n), the active power reference value of each unit;

8) judge the value size of n, if n≤30, then make n=n+1, jump to step 3), otherwise terminate this partial arithmetic.

The moment computing method that the active power reference value of each unit starts to act on are as follows:

When m in discrete wind series s (m) gets occurrence n, calculation of wind speed s (n) propagates into the time delay D of unit i from wind energy turbine set upwind border unit ^i,n, then when wind speed is s (n), the active power reference value of unit i the initiation moment for

$t_{ref}^{i,n}=t_N+D^{i,n}$ ⑨

Wind speed s (n) propagates into the time delay D of unit i from wind energy turbine set upwind border unit ^i,ncomputing method as follows:

1) wind presses wind direction propagation, on wind speed travel path, asks for the distance of the subpoint of impeller central on wind direction of wind energy turbine set upwind border unit impeller central and unit i, is designated as L;

2) D is calculated as follows ^i,n, and make i value travel through each unit of wind energy turbine set, each natural number in n value traversal 1≤n≤31,

$D^{i,n}=\frac{L}{s\left(n\right)}$ ⑩

Step 3: by each unit active power reference value and act on the moment, each unit active power reference value controlling curve under generating discrete wind series effect, each unit runs according to this controlling curve.

The generation method of unit active power reference value controlling curve is as follows: by i value, and n value is identical respectively with composition control amount its implication is in the moment, the active power reference value of unit i is to the i value determined, will value press the ascending arrangement of n, arrive between time, the active power reference value of unit i maintains constant, make i value travel through each unit of wind energy turbine set, each natural number in m value traversal 1≤m≤31, thus obtain wind energy turbine set each unit active power reference value controlling curve.Specifically can carry out as follows:

1) if wind energy turbine set unit adds up to N, the bivariate table of three capable 30 row of N is constructed, called after TableT, TableP and TalblePandT respectively;

2) the possible value of i and n is traveled through, will fill out and arrange at i-th row n-th of TableT;

3) the possible value of i and n is traveled through, will fill out and arrange at i-th row n-th of TableP;

4) i-th row n-th column element of TableT and i-th row n-th column element of TableP are complex as and the i-th row n-th being write on TalblePandT arranges;

5) take time as horizontal ordinate, active power reference value is that ordinate forms coordinate system, gets the i-th row data of TalblePandT, presses row number and sequentially fetches data successively from small to large, and demarcate each on coordinate system point, and, arrive between time, the active power reference value of unit i is maintained constant, thus obtain the active power reference value controlling curve of unit i; I traversal institute likely value, repeats this step, obtains each unit next control cycle active power reference value controlling curve.

Claims (6)

1., based on the active power of wind power field optimization method that wind speed sequential is decomposed, comprise the following steps:

Step one: the discrete wind series forecasting wind speed curve of next for wind energy turbine set control cycle being converted to chronologically 30 elements;

Step 2: air speed value corresponding for each for discrete wind series element is considered as constant natural wind speed, based on wake effect, calculates under each constant natural wind speed effect, the active power reference value of each unit respectively; Meanwhile, based on wake flow propagation delay, calculate respectively under each constant natural wind speed effect, each unit active power reference value starts the moment acted on;

Step 3: by each unit active power reference value and effect moment, each unit active power reference value controlling curve under generating discrete wind series effect, each unit runs according to this controlling curve.

2. the active power of wind power field optimization method decomposed based on wind speed sequential according to claim 1, it is characterized in that: in described step one, the conversion formula of discrete wind series s (m) is:

$s\left(m\right)=\frac{1}{T_s}{\∫}_{t_N+(m-1)T_s}^{t_N+{\mathrm{mT}}_s}{s}_c\left(t\right)dt$ ①

Wherein, m is natural number, and its span is 1≤m≤31, s _ct () is t predicting wind speed of wind farm value, t _nrepresent the initial time of next control cycle, T represents the duration that wind energy turbine set control cycle is corresponding, T _swind speed discretize interval time, T _s=T/30.

3. the active power of wind power field optimization method decomposed based on wind speed sequential according to claim 2, it is characterized in that: in described step 2, the calculation procedure of the active power reference value of each unit is as follows:

1) with the axial inducible factor a of unit i _ifor variable, then the power coefficient of unit i and thrust coefficient for:

$C_p^i=4{(1-a_i)}^2{a}_i$ ②

$C_T^i=4(1-a_i)a_i$ ③

2) m in discrete wind series s (m) gets occurrence n, makes n=1;

3) establish wind energy turbine set to be subject to constant natural wind speed s (n) effect, use v _irepresent the input wind speed of unit i, if v ₁be upwind border unit, unit i+1 is unit i leeward upwards First unit, 4. 5. calculates the input wind speed of unit i+1 by formula with formula:

v ₁＝s(n)④

$v_{i+1}=v_i\·\[1-(1-\sqrt{1-C_T^i}){\left(\frac{r_i}{r_i+kx}\right)}^2\]$ ⑤

Wherein, k is wind energy turbine set terrain rough factor, r _ifor unit i impeller radius, x is the distance that unit i hub centre and unit i+1 hub centre project on wind direction;

4) each unit active power is calculated as follows:

⑥

Wherein, P _irepresent the active power of unit i, v _ratedrepresent unit wind rating, v _in, v _cutrepresent the incision of unit, cut-out wind speed; ρ represents atmospheric density, represent the rated power of unit i;

5) active power of wind power field Optimized model is set up as follows:

$P_{all}=\underset{a_i}{max}\left(\underset{i=1}{\overset{N}{\Σ}}{P}_i\right)$

s.t.

v ₁＝s(n)

$v_{i+1}=v_i\·\[1-(1-\sqrt{1-C_T^i}){\left(\frac{r_i}{r_i+kx}\right)}^2\]$ ⑦

$0\≤a_i\≤\frac{1}{3}$

$0.2P_{rated}^i\≤P_i\≤P_{rated}^i$

Wherein, P _allrepresent active power of wind power field, i.e. wind energy turbine set each unit active power sum, N represents unit sum in wind energy turbine set, represent by optimizing a _ithe formula in bracket is made to obtain maximal value;

6) solution procedure 5) given by equation, obtain the optimization solution of unit i inducible factor, and substituted into formula 2.-5., draw the concrete value of unit i power coefficient, thrust coefficient, input wind speed;

7) unit i power coefficient value is substituted into following formula, then when to obtain nature wind speed be s (n), the active power reference value of unit i

$P_{ref}^{i,n}=\frac{1}{2}{\mathrm{\ρ\πr}}_i^2{v}_i^3{C}_p^i$ ⑧

Make i value travel through wind energy turbine set each unit, then when to obtain nature wind speed be s (n), the active power reference value of each unit;

8) judge the value size of n, if n≤30, then make n=n+1, jump to step 3), otherwise terminate computing.

4. the active power of wind power field optimization method decomposed based on wind speed sequential according to claim 3, it is characterized in that: in described step 2, the computing method that the active power reference value of each unit starts the moment acted on are as follows: when the m in discrete wind series s (m) gets occurrence n, calculation of wind speed s (n) propagates into the time delay D of unit i from wind energy turbine set upwind border unit ^i,n, then when wind speed is s (n), the active power reference value of unit i the initiation moment for

$t_{ref}^{i,n}=t_N+D^{i,n}$ ⑨。

5. the active power of wind power field optimization method decomposed based on wind speed sequential according to claim 4, it is characterized in that: starting in the moment computing method acted on of the active power reference value of each unit, wind speed s (n) propagates into the time delay D of unit i from wind energy turbine set upwind border unit ^i,ncomputing method as follows:

1) wind presses wind direction propagation, on wind speed travel path, asks for the distance of wind energy turbine set upwind border unit impeller central and the subpoint of unit i impeller central on wind direction, is designated as L;

2) D is calculated as follows ^i,n, and make i value travel through each unit of wind energy turbine set, each natural number in n value traversal 1≤n≤31,

$D^{i,n}=\frac{L}{s\left(n\right)}$ ⑩。

6. the active power of wind power field optimization method decomposed based on wind speed sequential according to claim 5, it is characterized in that: in described step 3, the generating mode of unit active power reference value controlling curve is as follows: by i value, and n value is identical respectively with composition control amount its implication is in the moment, the active power reference value of unit i is to the i value determined, will value press the ascending arrangement of n, arrive between time, the active power reference value of unit i maintains constant, thus obtain the active power reference value controlling curve of unit i; Make i value travel through each unit of wind energy turbine set, each natural number in m value traversal 1≤m≤31, thus obtain wind energy turbine set each unit active power reference value controlling curve.