CN103701143A - Energy storage configuration method for smoothing power fluctuation of wind and photovoltaic power storage system - Google Patents

Abstract

The invention discloses an energy storage configuration method for smoothing the power fluctuation of a wind and photovoltaic power storage system. The method comprises the following steps: firstly, fixing distributive generating system composition of stored energy to be distributed, definitizing historical generating data of various distributive generating equipment of the system, and a forming distributive generating power sample, secondly, performing discrete Fourier transform to power output sample data, thirdly, fixing a stabilizing target of the power fluctuation of the distributive generating system, fourthly, fixing a theoretical value required for energy storage power, fifthly, correcting the power requirement of an energy storage battery, and fixing final energy storage power, and sixthly, fixing the capacity requirement of the energy storage battery, and accomplishing the energy storage configuration. By configuring an energy storage system, distributive power output power fluctuation is smoothed through the operation characteristics that the system can be charged and discharged of the system, and as a distributive power source is accessed to a power grid, the stable operation of the power grid is guaranteed.

Description

The energy storage collocation method of level and smooth wind-light storage system power fluctuation

Technical field

The present invention relates to level and smooth wind-light storage system, relate in particular to a kind of energy storage collocation method of level and smooth wind-light storage system power fluctuation.

Background technology

At present, the intermittent distributed power sources such as wind-powered electricity generation, photovoltaic generation generate electricity and can access electrical network and can cause mains frequency fluctuation and then increase the assistant service burdens such as electrical network frequency modulation, pressure regulation, traffic control, cause the increase of operation of power networks cost; When power fluctuation exceeds electric power system peak modulation capacity scope, also will further cause power system frequency out-of-limit, the safe operation of serious threat electric power system.

In distributed generation system, mainly from energy storage device on the impact of distributed power source and the angle that maintains energy output and the load equilibrium of supply and demand, the configuration of energy-storage battery is mainly to take economy as target, layoutprocedure gathers the intermittence of considering distributed power source, the method existing institute in ripe software applies, the energy storage collocation method that the economy of take as just applied in homer software is target, its essence is that the intermittence for distributed power source generating compensates.Be characterized in that time scale is larger, the effect of energy-storage battery is equivalent to a controlled distributed power generation unit.Whether the drawback of prior art is mainly the economic performance of taking into account system only, cannot can stablize by taking into account system.

Summary of the invention

The energy storage collocation method that the object of this invention is to provide a kind of level and smooth wind-light storage system power fluctuation, can realize level and smooth to distributed power source output-power fluctuation, coordinates distributed power source access electrical network, guarantees the stable operation of electrical network.

The present invention adopts following technical proposals: a kind of energy storage collocation method of level and smooth wind-light storage system power fluctuation, comprises the following steps:

(1), determine the distributed generation system formation of energy storage to be joined, the history generating data of all kinds of distributed power generation equipment of the system that defines, form the power sample of distributed power generation, the generated output of the distributed power source on interconnection carries out that basis adds to being connected to and, form the gross power data of system distributed power source generating;

(2), power stage sample data is carried out to discrete Fourier transform;

(3), determine distributed generation system power fluctuation stabilize target;

(4), determine the theoretical value of energy storage power demand;

(5), revise the power demand of energy-storage battery, determine final energy storage power;

(6), determine the capacity requirement of energy-storage battery, complete energy storage configuration.

Described step is carried out discrete Fourier transform to power stage sample data in (2), obtains amplitude-frequency result S _gand f _g:

S _g=DFT(P _g)=[S _g[1],...,S _g[n],...,S _g[N _s]] ^T （1）

f _g=[f _g[1],...,f _g[n],...,f _g[N _s]] ^T

In formula, P _g=[P _g[1] ..., P _g[n] ..., P _g[N _s]] ^trepresent regenerative resource power stage sample data, P _g[n] represents n sampled point power output, and unit is kW; N _srepresent sampled point number; DFT (P _g) represent sample data P _gcarry out discrete Fourier transform;

S _g[n]=R _g[n]+I _g[n] i, represents n frequency f in Fourier transform result _gthe amplitude that [n] is corresponding, R _g[n], I _g[n] represents respectively real part and the imaginary part of amplitude; f _gfor with S _gfor corresponding column of frequencies vector;

f _g[n]=f _s(n-1)/N _s=(n-1)/(T _sN _s)（2）

In formula, f _s, T _sbe respectively sample data P _gsample frequency [Hz] and sampling period [s];

The frequency of the actual magnitude of original signal and correspondence thereof is respectively by column vector D _g,

represent:

In formula,

n is got in representative _s/ 2 integer part; D _g[j] represents j frequency f in spectrum analysis _gthe original signal actual magnitude size that [j] is corresponding;

Work as N _sduring for even number:

$D_g\left[j\right]=\left\{\begin{array}{cc}\sqrt{R_g^2\left[j\right]+I_g^2\left[j\right]}/N_s& j=1,N_s/2+1\\ 2\sqrt{R_g^2\left[j\right]+I_g^2\left[j\right]}/N_s& j=2,...,N_s/2\end{array}\right.---\left(4\right)$

Work as N _sduring for odd number:

Described step (3) is based on result of spectrum analysis, determines and meets the target power output of power stage fluctuation constraint and corresponding ESS compensation frequency range thereof, f _psrepresentative is according to result of spectrum analysis D _gdefinite compensation frequency range, f _ps1represent S _gin with Nyquist frequency f _nfor symmetry axis and f _pssymmetrical frequency range; Use S ₀=[S ₀[1] ..., S ₀[n] ..., S ₀[N _s]] ^tthe target power of representative after ESS compensation exported corresponding spectrum analysis complex result; Wherein, amplitude corresponding to compensation frequency range is set to 0, represents to have eliminated after compensating the power fluctuation of corresponding band, the amplitude outside compensation frequency range is constant,

$S_0\left[n\right]=\left\{\begin{array}{cc}0+0i& f_n\∈f_{\mathrm{ps}}U{f}_{\mathrm{ps}1}\\ S_g\left[n\right]& f_n\∉f_{\mathrm{ps}}U{f}_{\mathrm{ps}1}\end{array}\right.---\left(6\right).$

Described step (4) is specific as follows:

To S ₀carry out discrete fourier inverse transformation and can obtain the target power Output rusults P after ESS compensation ₀:

P ₀=IDFT(S ₀)=[P ₀[1],...,P ₀[n],...,P ₀[N _s]] ^T （7）

In formula, IDFT (S ₀) represent S ₀carry out discrete fourier inverse transformation; P ₀[n] represents the target output [kW] of n sampled point;

At T _epower fluctuation rate in time period is used

represent, its computing formula is as follows:

$F_{T_E}=\frac{P_{T_E}^{\max }-P_{T_E}^{\min }}{P_n}\×100\%---\left(8\right)$

In formula, P _nrepresent rated power [kW];

represent respectively T _emaximum and minimum output power [kW] in time period; Judge that whether target power output meets the demands, and needs to guarantee fluctuation ratio be no more than the upper limit of setting

$F_{T_E}\≤F_{T_E}^{\mathrm{up}}---\left(9\right).$

Described step (5) is specific as follows:

At definite ideal power target output P ₀afterwards, the power stage of required ESS is by column vector P _b0=[P _b0[1] ..., P _b0[n] ..., P _b0[N _s]] ^trepresent:

P _b0[n]=P ₀[n]-P _g[n] （10）

P _b0[n] can just can bear, and for just representing ESS electric discharge, is negative representative charging; The efficiency that ESS discharges and recharges a circulation is called ESS overall efficiency, uses η _eSrepresent; According to the power stage value of required ESS, consider the overall efficiency of ESS, determine the actual power that discharges and recharges of ESS, use P _b=[P _b[1] ..., P _b[n] ..., P _b[N _s]] ^trepresent:

$P_b\left[n\right]=\left\{\begin{array}{cc}{P}_{b0}\left[n\right]/{\mathrm{\&eta;}}_{\mathrm{ES},d}& P_{b0}\left[n\right]\&GreaterEqual;0\\ P_{b0}\left[n\right]g{\mathrm{\&eta;}}_{\mathrm{ES},c}& P_{b0}\left[n\right]<0\end{array}\right.---\left(11\right)$

In formula (11), η _{eS, c}and η _{eS, d}represent respectively ESS charge efficiency and discharging efficiency, if supposition ESS efficiency for charge-discharge is equal,

consider and discharge and recharge after power loss, when electric discharge, after the loss of ESS actual discharge power deduction, need to meet required discharge power requirement, its value is for required discharge power is divided by discharging efficiency; When charging, the actual charge power of ESS is the value after required charge power deduction charging loss, should be required charge power and is multiplied by charge efficiency;

Target power output after ESS compensation not only will meet power fluctuation and retrain, and also will guarantee that ESS can continuous and steady operation; , require in whole sample cycle for this reason, in ESS running, meet clean charge capacity, discharge electricity amount is zero, that is:

$\mathrm{\&Delta;E}=\underset{n=1}{\overset{\mathrm{Ns}}{\mathrm{\&Sigma;}}}\left(P_b\right[n]g{T}_s/3600)=0---\left(12\right)$

In formula, Ts represents the sampling period [s]; T _s/ 3600 represent chronomere's " second " convert be chronomere " hour ";

By P ₀whole translation downwards, aims of systems power stage translational movement is designated as Δ P, obtains, system power target output P after translation by iterative computation _a=[P _a[1] ..., P _a[n] ..., P _a[N _s]] ^trepresent:

P _a[n]=P ₀[n]-ΔP （13）

Corresponding to the aims of systems power stage P after translation _a, the power stage of required ESS is:

P _b0[n]=P _a[n]-P _g[n] （14）

Utilize formula (12) to obtain and consider that ESS discharges and recharges the actual performance number that discharges and recharges of ESS after power loss,

In whole sample data in the cycle, the actual power P that discharges and recharges of ESS obtaining _bthe maximum of absolute value is the maximum that ESS should possess and discharges and recharges power, that is ESS power-handling capability: P _eS0=max{|P _b[n] | }.

Described step (6) specifically comprises the following steps:

1) the ESS real output data based on definite, add up the ESS charge/discharge electricity amount of each sample point, can obtain different sampling instant ESS with respect to the energy hunting of initial condition, that is:

$E_{b,\mathrm{acu}}\left[m\right]=\underset{0}{\overset{m}{\mathrm{\&Sigma;}}}\left(P_b\right[m]g{T}_s/3600),m=0,...,N_s---\left(15\right)$

In formula, E _{b, acu}[m] represents ESS energy hunting with respect to initial condition m sampling instant, that is m (from the 0th to the m) sampling period before corresponding, ESS accumulative total charge-discharge energy sum [kWh];

2) for ESS energy hunting in the cycle in whole sample data, calculate the poor of ESS maximum, least energy, consider SOC restriction, obtain the capacity that ESS should possess, that is ESS rated capacity value E _eS0:

$E_{\mathrm{ES}0}=\frac{\max \left\{E_{b,\mathrm{acu}}\right[m\left]\right\}-\min \left\{E_{b,\mathrm{acu}}\right[m\left]\right\}}{C_{\mathrm{up}}-C_{\mathrm{low}}}---\left(16\right)$

In formula, C _upand C _lowrepresent respectively the constraint of ESS operation SOC upper and lower limit; C _upand C _lowin [0,1] interior value; Max{E _{b, acu}[m] }, min{E _{b, acu}[m] } represent respectively in the whole sample data cycle that ESS is with respect to minimum, the ceiling capacity of initial condition, max{E _{b, acu}[m] }-min{E _{b, acu}[m] } represented the absolute value of ESS ceiling capacity fluctuation in the whole sample data cycle.

The present invention proposes a kind of novel energy storage collocation method containing blower fan, photovoltaic distributed electricity generation system, from the angle of the compensation distributed power source randomness of exerting oneself, the difference of level and smooth power curve and the actual power curve of distributed power source of take is optimized stored energy capacitance as foundation.By configuration energy-storage system, rely on it can fill the operation characteristic that can put, realization is level and smooth to distributed power source output-power fluctuation, coordinates distributed power source access electrical network, guarantees the stable operation of electrical network.The method has separately carried out fully considering to the randomness of distributed power generation on the consideration intermittent basis of distributed power generation, realized the energy storage configuration of distributed generation system, can realize distributed power generation randomness is stabilized, can better coordinate distributed power source generating, optimize energy utilization rate, can improve the permeability of distributed power source.

Accompanying drawing explanation

Fig. 1 is flow chart of the present invention.

Embodiment

The invention discloses a kind of energy storage collocation method of level and smooth wind-light storage system power fluctuation, be mainly applicable to containing in photovoltaic, blower fan distributed electricity generation system, major function is by configuration energy-storage system, wind and light generating system power fluctuation to be stabilized.As shown in Figure 1, the configuration of energy-storage system is mainly divided into three links (six steps), and the first link is for optimizing the preparatory stage, is mainly historical data is searched and to historical data analysis, so that the carrying out of follow-up Optimization Work; The second link is energy-storage battery power fixed ring joint really, is mainly by the accumulate power of battery frequency range of expectation being analyzed, being determined the power demand of energy-storage battery; The 3rd link is accumulate energy content of battery demand, by energy-storage battery being discharged and recharged to determining of time frequency range, calculates the capacity requirement of energy-storage battery.

The present invention specifically comprises following six steps:

(1), determine the distributed generation system formation of energy storage to be joined, the history generating data of all kinds of distributed power generation equipment of the system that defines, form the power sample of distributed power generation, the generated output of the distributed power source on interconnection carries out that basis adds to being connected to and, form the gross power data of system distributed power source generating;

(2), power stage sample data is carried out to discrete Fourier transform;

Power stage sample data is carried out to discrete Fourier transform, obtain amplitude-frequency result S _gand f _g:

S _g=DFT(P _g)=[S _g[1],...,S _g[n],...,S _g[N _s]] ^T （1）

f _g=[f _g[1],...,f _g[n],...,f _g[N _s]] ^T

In formula, P _g=[P _g[1] ..., P _g[n] ..., P _g[N _s]] ^trepresent regenerative resource power stage sample data, P _g[n] represents n sampled point power output, and unit is kW; N _srepresent sampled point number; DFT (P _g) represent sample data P _gcarry out discrete Fourier transform;

S _g[n]=R _g[n]+I _g[n] i, represents n frequency f in Fourier transform result _gthe amplitude that [n] is corresponding, R _g[n], I _g[n] represents respectively real part and the imaginary part of amplitude; f _gfor with S _gfor corresponding column of frequencies vector;

f _g[n]=f _s(n-1)/N _s=(n-1)/(T _sN _s) （2）

In formula, f _s, T _sbe respectively sample data P _gsample frequency [Hz] and sampling period [s];

The actual magnitude of original signal (original signal refers to that initial data is corresponding with sample data, is the data before sample is chosen) and corresponding frequency thereof are respectively by column vector D _g,

represent:

In formula,

n is got in representative _s/ 2 integer part; D _g[j] represents j frequency in spectrum analysis, f _gthe original signal actual magnitude size that [j] is corresponding;

Work as N _sduring for even number:

$D_g\left[j\right]=\left\{\begin{array}{cc}\sqrt{R_g^2\left[j\right]+I_g^2\left[j\right]}/N_s& j=1,N_s/2+1\\ 2\sqrt{R_g^2\left[j\right]+I_g^2\left[j\right]}/N_s& j=2,...,N_s/2\end{array}\right.---\left(4\right)$

Work as N _sduring for odd number:

(3), determine distributed generation system power fluctuation stabilize target;

Based on result of spectrum analysis, determine and meet the target power output of power stage fluctuation constraint and corresponding ESS compensation frequency range thereof, f _psrepresentative is according to result of spectrum analysis D _gdefinite compensation frequency range, f _ps1represent S _gin with Nyquist frequency f _nfor symmetry axis and f _pssymmetrical frequency range; Use S ₀=[S ₀[1] ..., S ₀[n] ..., S ₀[N _s]] ^tthe target power of representative after ESS compensation exported corresponding spectrum analysis complex result; Wherein, amplitude corresponding to compensation frequency range is set to 0, represents to have eliminated after compensating the power fluctuation of corresponding band, the amplitude outside compensation frequency range is constant,

$S_0\left[n\right]=\left\{\begin{array}{cc}0+0i& f_n\&Element;f_{\mathrm{ps}}U{f}_{\mathrm{ps}1}\\ S_g\left[n\right]& f_n\&NotElement;f_{\mathrm{ps}}U{f}_{\mathrm{ps}1}\end{array}\right.---\left(6\right).$

(4), determine the theoretical value of energy storage power demand;

To S ₀carry out discrete fourier inverse transformation and can obtain the target power Output rusults P after ESS compensation ₀:

P ₀=IDFT(S ₀)=[P ₀[1],...,P ₀[n],...,P ₀[N _s]] ^T （7）

In formula, IDFT (S ₀) represent S ₀carry out discrete fourier inverse transformation; P ₀[n] represents the target output [kW] of n sampled point;

At T _epower fluctuation rate in time period is used represent, its computing formula is as follows:

$F_{T_E}=\frac{P_{T_E}^{\max }-P_{T_E}^{\min }}{P_n}\&times;100\%---\left(8\right)$

In formula, P _nrepresent rated power [kW];

represent respectively maximum in the TE time period and minimum output power [kW]; Judge that whether target power output meets the demands, and needs to guarantee fluctuation ratio

be no more than the upper limit of setting

$F_{T_E}\&le;F_{T_E}^{\mathrm{up}}---\left(9\right).$

(5), revise the power demand of energy-storage battery, determine final energy storage power;

At definite ideal power target output P ₀afterwards, the power stage of required ESS is by column vector P _b0=[P _b0[1] ..., P _b0[n] ..., P _b0[N _s]] ^trepresent:

P _b0[n]=P ₀[n]-P _g[n] （10）

P _b0[n] can just can bear, and for just representing ESS electric discharge, is negative representative charging; In actual ESS, in its charge and discharge process, have certain loss, the efficiency that ESS discharges and recharges a circulation is called ESS overall efficiency, uses η _eSrepresent.According to the power stage value of required ESS, consider the overall efficiency of ESS, can determine the actual power that discharges and recharges of ESS, use P _b=[P _b[1] ..., P _b[n] ..., P _b[N _s]] ^trepresent:

$P_b\left[n\right]=\left\{\begin{array}{cc}{P}_{b0}\left[n\right]/{\mathrm{\&eta;}}_{\mathrm{ES},d}& P_{b0}\left[n\right]\&GreaterEqual;0\\ P_{b0}\left[n\right]g{\mathrm{\&eta;}}_{\mathrm{ES},c}& P_{b0}\left[n\right]<0\end{array}\right.---\left(11\right)$

In formula (11), η _{eS, c}and η _{eS, d}represent respectively ESS charge efficiency and discharging efficiency, if supposition ESS efficiency for charge-discharge is equal,

consider and discharge and recharge after power loss, when electric discharge, after the loss of ESS actual discharge power deduction, need to meet required discharge power requirement, its value is for required discharge power is divided by discharging efficiency; When charging, the actual charge power of ESS is the value after required charge power deduction charging loss, should be required charge power and is multiplied by charge efficiency;

Target power output after ESS compensation not only will meet power fluctuation and retrain, and also will guarantee that ESS can continuous and steady operation; , require in whole sample cycle for this reason, in ESS running, meet clean charge capacity, discharge electricity amount is zero, that is:

$\mathrm{\&Delta;E}=\underset{n=1}{\overset{\mathrm{Ns}}{\mathrm{\&Sigma;}}}\left(P_b\right[n]g{T}_s/3600)=0---\left(12\right)$

In formula, T _srepresent the sampling period [s]; T _s/ 3600 represent chronomere's " second " convert be chronomere " hour ";

When utilizing ESS to compensate the power of given frequency range, due to what the power fluctuation of each frequency was compensated, it is complete cycle amount, if do not consider the loss that discharges and recharges of ESS, the required charge capacity of ESS should equal discharge electricity amount, that is to say that constraints (formula 12) will meet naturally.Yet, ESS actual efficiency η _eSbe less than 100%, now, the actual charge volume of ESS should be less than discharge capacity, i.e. Δ E>0.For guaranteeing that the output of system power target meets constraint (formula 12) and fluctuation ratio constraint (formula 9), can be by P ₀whole translation downwards, to do not changing power stage fluctuation ratio (by the known integral translation P of formula (formula 8) ₀can not change power stage fluctuation ratio) prerequisite under make Δ E=0.Aims of systems power stage translational movement is designated as Δ P, can obtain by iterative computation.System power target output P after translation _a=[P _a[1] ..., P _a[n] ..., P _a[N _s]] ^trepresent:

P _a[n]=P ₀[n]-ΔP （13）

Corresponding to the aims of systems power stage P after translation _a, the power stage of required ESS is:

P _b0[n]=P _a[n]-P _g[n] （14）

Utilize formula (12) to obtain and consider that ESS discharges and recharges the actual performance number that discharges and recharges of ESS after power loss,

In whole sample data in the cycle, the actual power P that discharges and recharges of ESS obtaining _bthe maximum of absolute value is the maximum that ESS should possess and discharges and recharges power, that is ESS power-handling capability: P _eS0=max{|P _b[n] | }.

(6), determine the capacity requirement of energy-storage battery, complete energy storage configuration;

1) the ESS real output data based on definite, add up the ESS charge/discharge electricity amount of each sample point, can obtain different sampling instant ESS with respect to the energy hunting of initial condition, that is:

$E_{b,\mathrm{acu}}\left[m\right]=\underset{0}{\overset{m}{\mathrm{\&Sigma;}}}\left(P_b\right[m]g{T}_s/3600),m=0,...,N---\left(15\right)$

In formula, E _{b, acu}[m] represents ESS energy hunting with respect to initial condition m sampling instant, that is m (from the 0th to the m) sampling period before corresponding, ESS accumulative total charge-discharge energy sum [kWh];

2) for ESS energy hunting in the cycle in whole sample data, calculate the poor of ESS maximum, least energy, consider SOC restriction, obtain the capacity that ESS should possess, that is ESS rated capacity value E _eS0:

$E_{\mathrm{ES}0}=\frac{\max \left\{E_{b,\mathrm{acu}}\right[m\left]\right\}-\min \left\{E_{b,\mathrm{acu}}\right[m\left]\right\}}{C_{\mathrm{up}}-C_{\mathrm{low}}}---\left(16\right)$

In formula, C _upand C _lowrepresent respectively the constraint of ESS operation SOC upper and lower limit; Ideally, C _up=1, C _low=0.While considering ESS actual motion, for fear of overcharging, cross film playback, ring ESS life-span, C _upand C _lowin [0,1] interior value; Max{E _{b, acu}[m] }, min{E _{b, acu}[m] } represent respectively in the whole sample data cycle that ESS is with respect to minimum, the ceiling capacity of initial condition, max{E _{b, acu}[m] }-min{E _{b, acu}[m] } represented the absolute value of ESS ceiling capacity fluctuation in the whole sample data cycle.

The present invention tackles distributed power source generating randomness, take and stabilize the energy storage collocation method that distributed power source fluctuation is target, considers the energy storage power demand correction of the factors such as energy-storage battery loss, conversion efficiency.Guaranteeing under the prerequisite of power grid operation, by configuration energy storage, improving the power fluctuation by blower fan or photovoltaic distributed power generation, improving the access capacity of distributed power source, fully using regenerative resource.

Claims (6)

1. the energy storage collocation method that level and smooth wind-light storage system power fluctuates, is characterized in that: comprise the following steps:

(1), determine the distributed generation system formation of energy storage to be joined, the history generating data of all kinds of distributed power generation equipment of the system that defines, form the power sample of distributed power generation, the generated output of the distributed power source on interconnection carries out that basis adds to being connected to and, form the gross power data of system distributed power source generating;

(2), power stage sample data is carried out to discrete Fourier transform;

(3), determine distributed generation system power fluctuation stabilize target;

(4), determine the theoretical value of energy storage power demand;

(5), revise the power demand of energy-storage battery, determine final energy storage power;

(6), determine the capacity requirement of energy-storage battery, complete energy storage configuration.

2. the energy storage collocation method of level and smooth wind-light storage system power fluctuation according to claim 1, is characterized in that: described step is carried out discrete Fourier transform to power stage sample data in (2), obtains amplitude-frequency result S _gand f _g:

S _g=DFT(P _g)=[S _g[1],...,S _g[n],...,S _g[N _s]] ^T （1）

f _g=[f _g[1],...,f _g[n],...,f _g[N _s]] ^T

In formula, P _g=[P _g[1] ..., P _g[n] ..., P _g[N _s]] ^trepresent regenerative resource power stage sample data, P _g[n] represents n sampled point power output, and unit is kW; N _srepresent sampled point number; DFT (P _g) represent sample data P _gcarry out discrete Fourier transform;

S _g[n]=R _g[n]+I _g[n] i, represents n frequency f in Fourier transform result _gthe amplitude that [n] is corresponding, R _g[n], I _g[n] represents respectively real part and the imaginary part of amplitude; f _gfor with S _gfor corresponding column of frequencies vector;

f _g[n]=f _s(n-1)/N _s=(n-1)/(T _sN _s)（2）

In formula, f _s, T _sbe respectively sample data P _gsample frequency [Hz] and sampling period [s];

The frequency of the actual magnitude of original signal and correspondence thereof is respectively by column vector D _g,

represent:

In formula,

n is got in representative _s/ 2 integer part; D _g[j] represents j frequency f in spectrum analysis _gthe original signal actual magnitude size that [j] is corresponding;

Work as N _sduring for even number:

$D_g\left[j\right]=\left\{\begin{array}{cc}\sqrt{R_g^2\left[j\right]+I_g^2\left[j\right]}/N_s& j=1,N_s/2+1\\ 2\sqrt{R_g^2\left[j\right]+I_g^2\left[j\right]}/N_s& j=2,...,N_s/2\end{array}\right.---\left(4\right)$

Work as N _sduring for odd number:

3. the energy storage collocation method that level and smooth wind-light storage system power according to claim 2 fluctuates, it is characterized in that: described step (3) is based on result of spectrum analysis, determine and meet the target power output of power stage fluctuation constraint and corresponding ESS compensation frequency range thereof, f _psrepresentative is according to result of spectrum analysis D _gdefinite compensation frequency range, f _ps1represent S _gin with Nyquist frequency f _nfor symmetry axis and f _pssymmetrical frequency range; Use S ₀=[S ₀[1] ..., S ₀[n] ..., S ₀[N _s]] ^tthe target power of representative after ESS compensation exported corresponding spectrum analysis complex result; Wherein, amplitude corresponding to compensation frequency range is set to 0, represents to have eliminated after compensating the power fluctuation of corresponding band, the amplitude outside compensation frequency range is constant,

$S_0\left[n\right]=\left\{\begin{array}{cc}0+0i& f_n\&Element;f_{\mathrm{ps}}U{f}_{\mathrm{ps}1}\\ S_g\left[n\right]& f_n\&NotElement;f_{\mathrm{ps}}U{f}_{\mathrm{ps}1}\end{array}\right.---\left(6\right).$

4. the energy storage collocation method that level and smooth wind-light storage system power according to claim 3 fluctuates, is characterized in that: described step (4) is specific as follows:

To S ₀carry out discrete fourier inverse transformation and can obtain the target power Output rusults P after ESS compensation ₀:

P ₀=IDFT(S ₀)=[P ₀[1],...,P ₀[n],...,P ₀[N _s]] ^T （7）

In formula, IDFT (S ₀) represent S ₀carry out discrete fourier inverse transformation; P ₀[n] represents the target output [kW] of n sampled point;

At T _epower fluctuation rate in time period is used

represent, its computing formula is as follows:

$F_{T_E}=\frac{P_{T_E}^{\max }-P_{T_E}^{\min }}{P_n}\&times;100\%---\left(8\right)$

In formula, P _nrepresent rated power [kW];

represent respectively T _emaximum and minimum output power [kW] in time period; Judge that whether target power output meets the demands, and needs to guarantee fluctuation ratio

be no more than the upper limit of setting

$F_{T_E}\&le;F_{T_E}^{\mathrm{up}}---\left(9\right).$

5. the energy storage collocation method that level and smooth wind-light storage system power according to claim 4 fluctuates, is characterized in that: described step (5) is specific as follows:

At definite ideal power target output P ₀afterwards, the power stage of required ESS is by column vector P _b0=[P _b0[1] ..., P _b0[n] ..., P _b0[N _s]] ^trepresent:

P _b0[n]=P ₀[n]-P _g[n] （10）

P _b0[n] can just can bear, and for just representing ESS electric discharge, is negative representative charging; The efficiency that ESS discharges and recharges a circulation is called ESS overall efficiency, uses η _eSrepresent; According to the power stage value of required ESS, consider the overall efficiency of ESS, determine the actual power that discharges and recharges of ESS, use P _b=[P _b[1] ..., P _b[n] ..., P _b[N _s]] ^trepresent:

$P_b\left[n\right]=\left\{\begin{array}{cc}{P}_{b0}\left[n\right]/{\mathrm{\&eta;}}_{\mathrm{ES},d}& P_{b0}\left[n\right]\&GreaterEqual;0\\ P_{b0}\left[n\right]g{\mathrm{\&eta;}}_{\mathrm{ES},c}& P_{b0}\left[n\right]<0\end{array}\right.---\left(11\right)$

In formula (11), η _{eS, c}and η _{eS, d}represent respectively ESS charge efficiency and discharging efficiency, if supposition ESS efficiency for charge-discharge is equal,

consider and discharge and recharge after power loss, when electric discharge, after the loss of ESS actual discharge power deduction, need to meet required discharge power requirement, its value is for required discharge power is divided by discharging efficiency; When charging, the actual charge power of ESS is the value after required charge power deduction charging loss, should be required charge power and is multiplied by charge efficiency;

Target power output after ESS compensation not only will meet power fluctuation and retrain, and also will guarantee that ESS can continuous and steady operation; , require in whole sample cycle for this reason, in ESS running, meet clean charge capacity, discharge electricity amount is zero, that is:

$\mathrm{\&Delta;E}=\underset{n=1}{\overset{\mathrm{Ns}}{\mathrm{\&Sigma;}}}\left(P_b\right[n]g{T}_s/3600)=0---\left(12\right)$

In formula, T _srepresent the sampling period [s]; T _s/ 3600 represent chronomere's " second " convert be chronomere " hour ";

By P ₀whole translation downwards, aims of systems power stage translational movement is designated as Δ P, obtains, system power target output P after translation by iterative computation _a=[P _a[1] ..., P _a[n] ..., P _a[N _s]] ^trepresent:

P _a[n]=P ₀[n]-ΔP （13）

Corresponding to the aims of systems power stage P after translation _a, the power stage of required ESS is:

P _b0[n]=P _a[n]-P _g[n] （14）

Utilize formula (12) to obtain and consider that ESS discharges and recharges the actual performance number that discharges and recharges of ESS after power loss,

In whole sample data in the cycle, the actual power P that discharges and recharges of ESS obtaining _bthe maximum of absolute value is the maximum that ESS should possess and discharges and recharges power, that is ESS power-handling capability: P _eS0=max{|P _b[n] | }.

6. the energy storage collocation method that level and smooth wind-light storage system power according to claim 5 fluctuates, is characterized in that: described step (6) specifically comprises the following steps:

1) the ESS real output data based on definite, add up the ESS charge/discharge electricity amount of each sample point, can obtain different sampling instant ESS with respect to the energy hunting of initial condition, that is:

$E_{b,\mathrm{acu}}\left[m\right]=\underset{0}{\overset{m}{\mathrm{\&Sigma;}}}\left(P_b\right[m]g{T}_s/3600),m=0,...,N---\left(15\right)$

In formula, E _{b, acu}[m] represents ESS energy hunting with respect to initial condition m sampling instant, that is m (from the 0th to the m) sampling period before corresponding, ESS accumulative total charge-discharge energy sum [kWh];

2) for ESS energy hunting in the cycle in whole sample data, calculate the poor of ESS maximum, least energy, consider SOC restriction, obtain the capacity that ESS should possess, that is ESS rated capacity value E _eS0:

$E_{\mathrm{ES}0}=\frac{\max \left\{E_{b,\mathrm{acu}}\right[m\left]\right\}-\min \left\{E_{b,\mathrm{acu}}\right[m\left]\right\}}{C_{\mathrm{up}}-C_{\mathrm{low}}}---\left(16\right)$

In formula, C _upand C _lowrepresent respectively the constraint of ESS operation SOC upper and lower limit; C _upand C _lowin [0,1] interior value; Max{E _{b, acu}[m] }, min{E _{b, acu}[m] } represent respectively in the whole sample data cycle that ESS is with respect to minimum, the ceiling capacity of initial condition, max{E _{b, acu}[m] }-min{E _{b, acu}[m] } represented the absolute value of ESS ceiling capacity fluctuation in the whole sample data cycle.