CN105353304A - Validation method of low voltage ride-through characteristic of electric model of wind turbine generator - Google Patents

Abstract

The present invention provides a validation method of a low voltage ride-through characteristic of an electric model of a wind turbine generator. The validation method comprises the following steps: determining a validation working condition requirement and external power grid model data of the low voltage ride-through characteristic validation of the electric model of the wind turbine generator; performing simulation and practical testing according to the validation working condition requirement, and obtaining the simulation result and the practical test result of the wind turbine generator; performing positive-sequence component calculation of the simulation result and the practical test result; performing data filtering of the positive-sequence component calculation result; dividing fault sections; and determining the accuracy of the electric model of the wind turbine generator according to the bias of the simulation result and the practical test result of the fault sections. According to the low voltage ride-through practical test result of the wind turbine generator, the separately sectional and multi-index bias determination of the simulation result of the electric model of the wind turbine generator is performed, the low voltage ride-through control characteristic with different types is fully considered, and the accuracy of the electric model of the wind turbine generator is determined.

Description

A kind of Wind turbines electrical model low voltage crossing property verification method

Technical field

The present invention relates to a kind of electric system simulation and the method in checking field, specifically relate to a kind of Wind turbines electrical model low voltage crossing property verification method.

Background technology

In the past decade Wind Power Development is rapid, worldwide becomes important electric power supply part.In order to improve the stability of electrical network, the technical requirement of wind energy turbine set access electrical network has been formulated in the electric system of various countries operation commercial city, and adopts the wind turbine model through checking, meets the degree that interconnection technology specifies carry out simulation analysis to wind energy turbine set.

In order to improve the grid stability of wind power integration, China has put into effect national grid company standard Q/GDW392-2009 " wind energy turbine set access electric power network technique regulation " in succession, Bureau of Energy's (2010) No. 433 literary compositions " the grid-connected detection management tentative method of Wind turbines ", standard GB/T/T19963-2011 " wind energy turbine set access power system technology regulation ", to Wind turbines and wind farm grid-connectedly propose specific requirement, also contains the electrical model to Wind turbines, the requirement of wind energy turbine set electrical model, require by simulation means assess this wind energy turbine set whether have interconnection technology regulation required by grid-connected characteristic.

Wind turbines manufacturing situation is complicated, market mainstream Wind turbines model is numerous, the configuration of its main parts size is varied, these bring very large pressure to the work of low voltage crossing Site Detection, and the tentative of " low voltage ride-through capability of wind turbine generator system compliance evaluation way " effectively alleviates this pressure.The method of assessment is carried out model accuracy verification, is guaranteed model accuracy after requiring that wind turbine model has been set up.The Wind turbines electrical model qualified according to verification evaluates low voltage ride through characteristic.

Therefore, need to provide a kind of verification method accurately carrying out Wind turbines electrical model low voltage crossing characteristic.

Summary of the invention

For overcoming above-mentioned the deficiencies in the prior art, the invention provides a kind of Wind turbines electrical model low voltage crossing property verification method.

Realizing the solution that above-mentioned purpose adopts is:

A kind of Wind turbines electrical model low voltage crossing property verification method, its improvements are: said method comprising the steps of:

I, determine the data of the external electrical network model verifying working condition requirement and described Wind turbines electrical model low voltage crossing property verification;

II, carry out respectively emulating and actual test according to described checking working condition requirement;

III, positive-sequence component computing is carried out to the result of Step II;

IV, data filtering is carried out to the result of positive-sequence component computing;

V, division fault section;

VI, determine the accuracy of described Wind turbines electrical model according to the described reality test of fault section and the deviation of result of described emulation.

Further, described detection working condition requirement comprises high-power output state and miniwatt output state;

Described high-power output state comprises Voltage Drop in three-phase symmetrical fault and two-phase unbalanced fault situation respectively to (0.75 ± 0.05) U _n, (0.50 ± 0.05) U _n, (0.35 ± 0.05) U _n, (0.20 ± 0.05) U _noperating mode;

Described miniwatt output state comprises Voltage Drop in three-phase symmetrical fault and two-phase unbalanced fault situation respectively to (0.75 ± 0.05) U _n, (0.50 ± 0.05) U _n, (0.35 ± 0.05) U _n, (0.20 ± 0.05) U _noperating mode.

Further, the data of described external electrical network model comprise equivalent electric pessimistic concurrency control parameter and Voltage Drop device parameter;

Described equivalent electric pessimistic concurrency control parameter comprises: equivalent electrical network voltage U _gwith equivalent electrical network impedance Z _g;

Described Voltage Drop device parameter comprises: current-limiting impedance Z ₁resistance and reactance value, short-circuit impedance Z ₂resistance and reactance value, and switch S ₁and S ₂action sequence.

Further, described external electrical network model comprises the equivalent electrical network voltage U connected successively _g, equivalent electrical network impedance Z _g, current-limiting impedance Z ₁, wind turbine transformer and Wind turbines; Described current-limiting impedance Z ₁with switch S ₁parallel connection, short-circuit impedance Z ₂be connected to described current-limiting impedance Z ₁and between wind turbine transformer, described short-circuit impedance Z ₂pass through switch S ₂ground connection.

Further, the result of described Step II comprises voltage, electric current and the power between the wind turbine transformer of emulation and actual test and Wind turbines.

Further, described step V comprises the following steps:

S501, voltage data according to reality test, A period of three periods, B period, C period before the data sequence of actual testing and simulation is divided into corresponding fault, between age at failure, after fault;

S502, according to the response characteristic of active power described B period and described C period to be divided between the interval and steady-state zone of transient state;

S503, according to the response characteristic of reactive current described B period and described C period to be divided between the interval and steady-state zone of transient state.

Further, in described step S502 and S503, when considering that current-limiting impedance affects, between the transient state interval that the described C period comprises between described transient state interval, described steady-state zone, current-limiting impedance causes and the steady-state zone that current-limiting impedance causes.

Further, the deviation in described step VI comprises mean deviation, mean absolute deviation, maximal deviation sum weighted mean absolute deviation;

Day part transient state interval calculates described mean deviation and described mean absolute deviation respectively, calculates described mean deviation, described mean absolute deviation and described maximum deviation between day part steady-state zone respectively.

Compared with prior art, the present invention has following beneficial effect:

1, method provided by the invention with the actual test result of LVRT Capability of Wind Turbine Generator for foundation, the deviation of the simulation result of Wind turbines electrical model being carried out to by stages multi objective judges, be beneficial to the low voltage crossing control characteristic taking into full account different type of machines, accurately judge Wind turbines electrical model accuracy;

2, method provided by the invention inputs for data with three-phase voltage, three-phase current, carries out error calculation to its positive-sequence component result of calculation, more meets Wind turbines transient characterisitics examination demand;

3, when method provided by the invention divides fault section, taken into full account that the low voltage crossing characteristic of different type of machines is different, automatically can carry out the judgement in transient state interval according to test data fluctuation, avoid false judgment;

4, method provided by the invention, has taken into full account test data fluctuation situation, has proposed different interval deviation wire examination methods, the consistance of comprehensive evaluation test result and simulation result.

Accompanying drawing explanation

Fig. 1 is the external electrical network structural representation of LVRT Capability of Wind Turbine Generator of the present invention test;

Fig. 2 is the fault subregion schematic diagram in the present embodiment.

Embodiment

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in further detail.

The invention provides a kind of Wind turbines electrical model low voltage crossing property verification method, the method is by carrying out consistency analysis to the emulated data of Wind turbines electrical model and actual test data, positive-sequence component calculating and data filtering are carried out to input data, carries out fault section division; Deviation calculating is carried out to the test in each fault section divided and simulation process data, according to above-mentioned deviation result of calculation examination Wind turbines electrical model precision, ensures the accuracy of simulation of Wind turbines electrical model.

The method specifically comprises the following steps:

Step one, determine the external electrical network model data of the realistic model verifying working condition requirement and described Wind turbines electrical model low voltage crossing property verification;

Step 2, carry out the emulation of described Wind turbines electrical model and the actual test of corresponding Wind turbines according to described checking working condition requirement, obtain the simulation result of wind turbine transformer low-pressure side or on high-tension side voltage, electric current and power and actual test result;

Step 3, positive-sequence component computing is carried out to described actual test result and described simulation result;

Step 4, data filtering is carried out to the result of positive-sequence component computing;

Step 5, fault section divide;

Step 6, determine the accuracy of described Wind turbines electrical model according to the described actual test result of each fault section and the deviation of described simulation result.

In step one, each working condition requirement when checking working condition requirement is for detecting, specifically comprises two kinds of active power output states: high-power output state (P>0.9P _n) and miniwatt output state (0.1P _n≤ P≤0.3P _n).

Under high-power output state and miniwatt output state, under fault type includes three-phase symmetrical fault and two-phase unbalanced fault situation, Voltage Drop is respectively to (0.75 ± 0.05) U _n, (0.50 ± 0.05) U _n, (0.35 ± 0.05) U _n, (0.20 ± 0.05) U _n, to sum up have 16 kinds of operating modes.

The acquisition methods of the modeling data of the realistic model of Wind turbines electrical model low voltage crossing property verification is for obtain modeling data by data collection point.

As shown in Figure 1, Fig. 1 is the external electrical network structural representation that LVRT Capability of Wind Turbine Generator of the present invention detects to said external electric network model; Method of the present invention only relate to outside equivalent electrical network and Voltage Drop device model arrange, outside equivalent electrical network and Voltage Drop mounted cast are universal model, for different manufacturers Wind turbines electrical model or different operating mode only need change optimum configurations.

The equivalent circuit diagram of this low-voltage ride through detection system comprises the equivalent electrical network voltage U connected successively _g, equivalent electrical network impedance Z _g, current-limiting impedance Z ₁, wind turbine transformer and Wind turbines, above-mentioned current-limiting impedance Z ₁paralleling switch S ₁, current-limiting impedance Z ₁short-circuit impedance Z is connected with between wind turbine transformer ₂, this short-circuit impedance Z ₂pass through switch S ₂ground connection.

Data collection point comprises three, is respectively equivalent electric network impedance and current-limiting impedance Z ₁between a bit (collection point one), short-circuit impedance Z ₂access point and wind turbine transformer between a bit (collection point two), between wind turbine transformer and Wind turbines a bit (collection point three).

According to real system, obtain the data of collection point one, collection point two, collection point three respectively; Comprise:

1, the voltage of collection point one, electric current and test period wind speed is obtained;

2, obtain the test datas such as the voltage of collection point two, electric current and generator speed, wind energy conversion system rotating speed, pitch angle to choose according to the demand of modelling verification;

3, obtain the test datas such as the voltage of collection point three, electric current and generator speed, wind energy conversion system rotating speed, pitch angle to choose according to the demand of modelling verification.

Simulation modeling desired data comprises equivalent electric pessimistic concurrency control parameter and Voltage Drop device parameter.

Equivalent electric pessimistic concurrency control parameter comprises: equivalent electrical network voltage U _g, voltage sag generator grid side access point capacity of short circuit S _k, angle of impedance Ψ or impedance ratio X/R.

Voltage Drop device parameter comprises: current-limiting impedance Z ₁resistance value and reactance value, short-circuit impedance Z ₂resistance value and reactance value and switch S ₁and S ₂action sequence.

Above-mentioned equivalent electrical network voltage U _gdefining method be: Closing Switch S ₁, obtain the magnitude of voltage positive-sequence component of collection point three.

Voltage sag generator grid side access point capacity of short circuit S _kdefining method be: voltage sag generator grid side access point capacity of short circuit u _gfor equivalent line voltage, Z _gfor equivalent electric network impedance.In the present embodiment, its defining method comprises the following steps:

(1) no load test obtains the voltage U at collection point three place ₃;

(2) load testing obtains the voltage at collection point three place is U ₃' and electric current I ₃;

(3) system impedance Z=(U ₃-U ₃')/I ₃, power system capacity S _k=U ₃ ²/ Z.

Above-mentioned angle of impedance Ψ or impedance ratio X/R is arranged by local grid condition empirical value.

Required device parameter when above-mentioned Voltage Drop device parameter obtains modeling according to the equipment in real system.

In step 2, carry out the emulation of described Wind turbines electrical model according to described working condition requirement, obtain emulated data.

Emulated data comprises: the test data needed for Wind turbines electrical model low voltage crossing property verification method and emulated data, comprises three-phase voltage, three-phase current and test data, emulated data and the sampling time of data for verifying respectively.

Above-mentioned three-phase voltage and three-phase current refer to the result of wind turbine transformer low-pressure side or on high-tension side voltage, electric current and power.

In step 3, positive-sequence component computing is carried out to the test data inputted and emulated data, be specially and the three-phase voltage in testing and emulation result, three-phase current data are carried out positive-sequence component calculating according to IEC61400-21 standard-required, obtain the fundamental positive sequence of the line voltage of test data and emulated data, active power, reactive power and reactive current.

In step 4, sampling time (can determine according to the sampling time of emulated data, be set to 1ms or 20ms in the present embodiment) positive-sequence component to test data and emulated data according to model checking carries out digital filtering, line time sequence synchronization of going forward side by side.

Digital filtering refers to use certain algorithm to process digital signal by digital device, the signal of certain frequency range is carried out filtering, obtains new signal.Common equipment is digital filter, and digital filter is divided into two large divisions: classical filter device and Modern Filter.

Classical filter device: assuming that useful component in input signal x (n) and wish that filtering composition lays respectively at different frequency bands, by a linear system, filtering is carried out to noise, if the mutual aliasing of the frequency spectrum of noise and signal, then classical filter device can not get the requirement of filtering.Usually Hi-pass filter is had, low-pass filter, bandpass filter, rejection filter.

Modern Filter: go out useful signal and noise signal from containing noisy Signal estimation.The method is that signal and noise itself are all considered as random signal, utilizes its statistical nature, and as autocorrelation function, cross correlation function, auto-power spectrum, cross-power spectrum etc. are guided out the algorithm for estimating of signal, then utilize digital device to realize.Mainly contain Wiener filtering, Kalman filtering, the digital filters such as auto adapted filtering.

In step 5, divide fault section.With actual test data for foundation, subregion is carried out to failure process, as shown in the fault subregion schematic diagram in Fig. 2 the present embodiment.

I, according to test voltage data, the data sequence of testing and simulation is divided into A (before fault), B (between age at failure), three periods of C (after fault);

1) before Voltage Drop, 1s starts the A period;

2) Voltage Drop is to 0.9U _nthe front 20ms in moment terminates the A period, and the B period starts;

3) the front 20ms of fault clearance start time terminates the B period, and the C period starts

4), after fault clearance, the 1s after Wind turbines active power starts stable output terminates the C period.

II, response characteristic according to active power, B, C period is divided between transient state interval and steady-state zone, wherein the B period is divided into B1_a (transient state) and B2_a (stable state) interval, the C period is divided into C1_a (transient state), C2_a (stable state) (if consider the impact of current-limiting impedance, also comprising C3_a (transient state that current-limiting impedance causes) and C4_a (stable state)) interval.

1) start time of B period is the beginning in B1_a transient state interval

2) general with the end of 100ms after Voltage Drop for B1_a transient state interval, if transient state process can not at this moment in terminate, then with the fluctuation of active power enter its mean value in this period ± 10% scope in the rear 20ms in moment be the end in B1_a and B1_r transient state interval.

3) start time of C period is the beginning in C1_a transient state interval;

4) generally with the finish time of the 500ms after fault clearance for the interval C1_a of active power transient state; The end of rear 250ms for the interval C3_a of transient state is exited with current-limiting impedance.If transient state process can not terminate within the above-mentioned time, then respectively with the fluctuation of active power and reactive current enter its mean value in this period ± 10% scope in the rear 20ms in moment be the end in C1_a, C3_a transient state interval.

III, response characteristic according to reactive current, B, C period is divided between transient state interval and steady-state zone, wherein the B period is divided into B1_r (transient state) and B2_r (stable state) interval, the C period is divided into C1 (transient state), C2 (stable state) (if consider the impact of current-limiting impedance, also comprising C3 (transient state that current-limiting impedance causes) and C4 (stable state)) interval.

1) start time of B period is the beginning in B1_r transient state interval.

2) general with the end of 100ms after Voltage Drop for B1_r transient state interval.If transient state process can not terminate in the above-mentioned time, then respectively with the fluctuation of active power, reactive current enter its mean value in this period ± 10% scope in the rear 20ms in moment be the end in B1_r transient state interval.

3) start time of C period is the beginning in C1_r transient state interval;

4) generally with the finish time of the 100ms after fault clearance for the interval C1_r of active power transient state; The end of rear 250ms for the interval C3_r of transient state is exited with current-limiting impedance.If transient state process can not terminate within the above-mentioned time, then respectively with the fluctuation of reactive current enter its mean value in this period ± 10% scope in the rear 20ms in moment be the end in C1_r, C3_r transient state interval.

In step 6, by calculating the deviation between test data and emulated data, the order of accuarcy of examining model.The electric parameters that testing and simulation deviation calculates is: voltage, active power, reactive power and reactive current.

Testing and simulation deviation comprises mean deviation, mean absolute deviation, maximum deviation and weighted mean absolute deviation.Wherein, day part transient state interval calculates mean deviation and mean absolute deviation respectively, calculates mean deviation, mean absolute deviation and maximum deviation respectively between steady-state zone.

In the present embodiment, use X _sand X _mrepresent the emulated data of above electric parameters and the perunit value of test data fundamental positive sequence respectively.K _startand K _endfirst and sequence number corresponding to last emulated data, test data when representing calculation deviation respectively.

1, mean deviation:

Calculate the arithmetic mean of test data and emulated data fundamental positive sequence difference, and get its absolute value, use F ₁represent.

$F_1=\left|\frac{\underset{i=K_{\mathrm{Start}}}{\overset{K_{\mathrm{End}}}{\mathrm{\Σ}}}(X_M\left(i\right)-X_S\left(i\right))}{K_{\mathrm{End}}-K_{\mathrm{Start}}+1}\right|---\left(1\right)$

2, mean absolute deviation:

Calculate the arithmetic mean of the absolute value of test data and emulated data fundamental positive sequence difference, use F ₂represent.

$F_2=\left|\frac{\underset{i=K_{\mathrm{Start}}}{\overset{K_{\mathrm{End}}}{\mathrm{\Σ}}}\left|(X_M\left(i\right)-X_S\left(i\right))\right|}{K_{\mathrm{End}}-K_{\mathrm{Start}}+1}\right|---\left(2\right)$

3, maximum deviation

Calculate the maximal value of the absolute value of test data and emulated data fundamental positive sequence difference, use F ₃represent.

$F_3={\max }_{i=K_{\mathrm{Start}}...K_{\mathrm{End}}}\left(\right|X_M\left(i\right)-X_S\left(i\right)\left|\right)---\left(3\right)$

4, weighted mean absolute deviation

Calculate active power, reactive power, reactive current respectively in the mean absolute deviation of A, B, C period, with F _aP, F _bP, F _cP, F _aQ, F _bQ, F _cQ, F _aIQ, F _bIQ, F _cIQrepresent.

For the mean absolute deviation FBP of B period active power, K _startand K _endrepresent the sequence number that B period data sequence first is corresponding with last data respectively.Be calculated as follows:

$F_{\mathrm{BP}}=\frac{\underset{i=K_{\mathrm{Start}}}{\overset{K_{\mathrm{End}}}{\mathrm{\Σ}}}\left|(P_M\left(i\right)-P_S\left(i\right))\right|}{K_{\mathrm{End}}-K_{\mathrm{Start}}+1}---\left(4\right)$

Wherein, P _m(i), P _si () represents active power, reactive power or reactive current in test data and emulated data respectively;

The mean absolute deviation of day part is weighted on average, obtains the weighted mean absolute deviation of whole process.Three interval weights are respectively:

---A (before fault): 10%

---B (between age at failure): 60%

---C (after fault): 30%

Weighted mean absolute deviation is calculated as follows for active power:

F _{G_P}＝0.1F _AP+0.6F _BP+0.3F _CP(5)

Finally should be noted that: above embodiment is only for illustration of the technical scheme of the application but not the restriction to its protection domain; although with reference to above-described embodiment to present application has been detailed description; those skilled in the art still can carry out all changes, amendment or equivalent replacement to the embodiment of application after reading the application; but these change, revise or be equal to replacement, all applying within the claims awaited the reply.

Claims (8)

1. a Wind turbines electrical model low voltage crossing property verification method, is characterized in that: said method comprising the steps of:

I, determine the data of the external electrical network model verifying working condition requirement and described Wind turbines electrical model low voltage crossing property verification;

II, carry out respectively emulating and actual test according to described checking working condition requirement;

III, positive-sequence component computing is carried out to the result of Step II;

IV, data filtering is carried out to the result of positive-sequence component computing;

V, division fault section;

VI, determine the accuracy of described Wind turbines electrical model according to the described reality test of fault section and the deviation of result of described emulation.

2. the method for claim 1, is characterized in that: described detection working condition requirement comprises high-power output state and miniwatt output state;

Described high-power output state comprises Voltage Drop in three-phase symmetrical fault and two-phase unbalanced fault situation respectively to (0.75 ± 0.05) U _n, (0.50 ± 0.05) U _n, (0.35 ± 0.05) U _n, (0.20 ± 0.05) U _noperating mode;

Described miniwatt output state comprises Voltage Drop in three-phase symmetrical fault and two-phase unbalanced fault situation respectively to (0.75 ± 0.05) U _n, (0.50 ± 0.05) U _n, (0.35 ± 0.05) U _n, (0.20 ± 0.05) U _noperating mode.

3. the method for claim 1, is characterized in that: the data of described external electrical network model comprise equivalent electric pessimistic concurrency control parameter and Voltage Drop device parameter;

Described equivalent electric pessimistic concurrency control parameter comprises: equivalent electrical network voltage U _gwith equivalent electrical network impedance Z _g;

Described Voltage Drop device parameter comprises: current-limiting impedance Z ₁resistance and reactance value, short-circuit impedance Z ₂resistance and reactance value, and switch S ₁and S ₂action sequence.

4. the method for claim 1, is characterized in that: described external electrical network model comprises the equivalent electrical network voltage U connected successively _g, equivalent electrical network impedance Z _g, current-limiting impedance Z ₁, wind turbine transformer and Wind turbines; Described current-limiting impedance Z ₁with switch S ₁parallel connection, short-circuit impedance Z ₂be connected to described current-limiting impedance Z ₁and between wind turbine transformer, described short-circuit impedance Z ₂pass through switch S ₂ground connection.

5. the method for claim 1, is characterized in that: the result of described Step II comprises voltage, electric current and power between the wind turbine transformer of emulation and actual test and Wind turbines.

6. the method for claim 1, is characterized in that: described step V comprises the following steps:

S501, voltage data according to reality test, A period of three periods, B period, C period before the data sequence of actual testing and simulation is divided into corresponding fault, between age at failure, after fault;

S502, according to the response characteristic of active power described B period and described C period to be divided between the interval and steady-state zone of transient state;

S503, according to the response characteristic of reactive current described B period and described C period to be divided between the interval and steady-state zone of transient state.

7. method as claimed in claim 6, it is characterized in that: in described step S502 and S503, when considering that current-limiting impedance affects, between the transient state interval that the described C period comprises between described transient state interval, described steady-state zone, current-limiting impedance causes and the steady-state zone that current-limiting impedance causes.

8. the method for claim 1, is characterized in that: the deviation in described step VI comprises mean deviation, mean absolute deviation, maximal deviation sum weighted mean absolute deviation;

Day part transient state interval calculates described mean deviation and described mean absolute deviation respectively, calculates described mean deviation, described mean absolute deviation and described maximum deviation between day part steady-state zone respectively.