METHOD FOR DETERMINING A DROOP RESPONSE PROFILE OF AN ELECTRICAL MACHINE CONNECTED TO AN ELECTRICAL GRID

Abstract

A method for determining a droop response profile of a rotating electrical machine supplying electricity to an electrical network having a network frequency varying on either side of a nominal frequency, wherein a measured rotational speed value and droop response parameters are defined. The droop response profile is a graph centered on the coordinates ([X5; Y5]) of an origin point between 99% and 101% of the measured speed value and defined by at least two coordinate points ([X4,Y4], [X0, Y0]) in the case of over-speed, each of the points having as abscissa a speed value as a percentage of the measured speed value, and for ordinates a filtered speed value as a percentage of the measured speed value modulated by at least one of the droop response parameters.

Description

TECHNICAL FIELD

The present application relates generally to rotating machines generating electricity in order to satisfy the electricity requirements of an electrical network and more particularly relates to the control of such rotating machines.

BACKGROUND OF THE INVENTION

An electrical network must ensure a constant balance between electrical consumption and electrical generation. Thus, increasing electrical consumption results in a drop in the frequency of the electrical network. Conversely, a drop in electrical consumption results in increasing the frequency of the electrical network.

In order to maintain a constant balance between electrical consumption and generation, the output of the power generating groups may be regulated to maintain the frequency of the electrical network at, for example, around 50 Hz or so. The power output provided by each group of generators producing electricity may be defined by its droop. Specifically, droop may be defined as the ratio between power output variation and frequency variation.

In addition, the use of renewable energy also affects the stability of the electrical grid. Thus, the power generating groups may be required to modify their response profile in droop to the frequency variations of the electrical network. Also, there may be certain electrical networks requiring different responses from the power generating groups when the electrical network is operating in under or over frequency. The droop response profile of such electrical production may be called an “asymmetrical droop response profile.”

In order to limit the frequency instabilities generated by network compensating operations, it may be necessary to define a dead band within the group of electrical production that may not provide compensation to maintain the frequency, in spite of a continuous variation in the frequency of the electrical network. The dead band range may be determined by either the electrical producer or by the implementing rules for the electrical network defined by a transport network administrator (GRT) or by a transmission system operator (TSO). The administrator of a transport network also may define the parameters of an operating profile of the generator group such as the behavior at the exit of dead band, the droop for the group of electrical production, or a droop limiter.

As an example, U.S. Pat. No. 6,118,187 describes a procedure for implement a dynamic dead band in order to manage a dynamic frequency in an electrical network in terms of frequency and amplitude. In addition, U.S. Patent Publication No. 2014/0260293 describes a control device for a gas turbine, including a system for droop response configured to detect one or several operating features in a turbine. For this purpose, the control device may include a multi-variable correction method based on operational characteristics such as a derivation of the load dependent on the percentage of the speed, the percentage of the turbine frequency, and the derivation of the ambient temperature at the intake of the turbine compressor. The correction method thus may generate a series of correction factors for the droop response that make it possible to produce a graph of the behavior of the turbine with a correction on the ambient temperature as a function of the input temperature of the turbine compressor.

However, the known methods for configuring the droop response of a turbine may not allow for the automatic integration of several functions such as the dead band, the droop of the electrical generating group, the output of the dead band, or limiting the droop response in order to determine a response profile of the turbine to variations in speed. A value of 100% of speed may correspond to about 50 Hz or 60 Hz depending on the country.

Thus, an object of the present application is to remedy the aforementioned drawbacks and to propose a method of defining a static response profile of an electrical generation group capable of responding to the frequency variations of the electrical network.

SUMMARY OF THE INVENTION

The present application relates to a method of determining a response profile in droop or a speed profile of a rotating electrical machine supplying electricity to an electrical network. A network frequency may vary on either side of a nominal frequency in which a measured value (Vm) of the speed of rotation of the rotating machine corresponding to the image of the frequency of the electrical network and the response parameters in dependence of the measured speed value are defined. The static response profile may be a graph centered on the coordinates of an origin point between 99% and 101% of the measured speed, preferably equal to 100% of the measured speed, and defined by at least two points in the case of under-speed and by at least two points in the case of over-speed. Each of the points may have speed value as a percentage of the measured speed and for ordinates a filtered speed value as a percentage of the measured speed modulated by at least one of the droop response parameters. The value of the filtered speed may affect the fuel control loop.

The parameters may include at least the value of the high dead band and the low dead band on either side of the original coordinate point, the value of the low, the median, and the high droop of the rotating machine, the value of the low and high limiter droop, at least one dead band output mode, and the value of the low and high breaking points of the nonlinear droop. Advantageously, the coordinates of a first point may be calculated in the case of under-speed, corresponding to the low dead band, having as the abscissa equal to the subtraction of 100% of the measured speed value with the low dead band, and ordinates equal to 100% of the measured speed.

To define the dead band output, the value of a gain of the median droop and the value of a low droop gain may be calculated. The gain of the droop may correspond to a ratio between the intrinsic droop of the rotating machine, for example 4%, divided by the desired droop. For example, for a desired droop of 4% the corresponding gain may be 1 (4%/4%). Thus, for a real speed delta measured by 0.2% at the dead band output, the filtered delta may be 0.2%. Moreover, for a desired droop of 2%, the corresponding gain may be 2 (4%/2%). Thus, for a measured speed delta of 0.2% at the dead band output, the filtered delta may be 0.4%. For example, the coordinates of a second point may be calculated in the case of under-speed, corresponding to the output of the dead band, as a function of the dead band output mode, low droop limiter value, low dead band value, median droop gain, and low breaking point value.

Advantageously, the coordinates of a third point may be calculated in the case of under-speed, corresponding to the low breaking point of the non-linear droop, as a function of the coordinates of the second point, the value of the low breakpoint, the value of the low droop limiter, and median droop gain. Advantageously, the coordinates of a fourth point may be calculated in the case of under-speed, corresponding to the low droop limiter, as a function of the coordinates of the third point, the value of the low droop limiter, and the low droop gain. Advantageously, the coordinates of a fifth point may be calculated in the case of under-speed, corresponding to the low limit point of the response profile, as a function of the coordinates of the fourth point and of the value of the low droop limiter.

According to another embodiment, the coordinates of a first point may be calculated in the case of an over-speed corresponding to the high dead band and having on the abscissa equal to the addition of 100% of the measured speed with the value of the band 100% of the measured speed. To define the dead band output, the value of a high droop gain corresponding to the ratio between the intrinsic droop of the machine and the desired droop, for example 4%, may be calculated. For example, the coordinates of a second point may be calculated in the case of over-speed, corresponding to the output of the dead band, depending on the dead band output mode, the high droop limiter value, the value of the high dead band, the high droop gain, and the value of the high breakpoint.

Advantageously, the coordinates of a third point may be calculated in the case of over-speed, corresponding to the high breaking point of the non-linear droop, as a function of the coordinates of the second point, the value of the high break point, the value of the high droop limiter, and the median droop gain. Advantageously, the coordinates of a fourth point may be calculated in the case of over-speed, corresponding to the high droop limiter, as a function of the coordinates of the third point, of the value of the high droop limiter, and of the high droop gain. Advantageously, the coordinates of a fifth point may be calculated in the case of over-speed, corresponding to the high limit point of the response profile, as a function of the coordinates of the fourth point, and of the value of the high droop limiter.

The value of the low dead band may be, for example, between 0.02% and 6% of the measured speed value. The value of the high dead band may be, for example, between 0.02% and 1% of the measured speed value. At least one of the values of the median droop, the low droop, and the high droop may be, for example, between 2% and 20% of the measured speed value. At least one of the values of the low and high break points of the non-linear droop may be, for example, between 0% and 10% of the measured speed value. The value of the low droop limiter may be, for example, between 96% and 100% of the filtered speed value. The value of the high droop limiter may be, for example, between 100% and 104% of the filtered speed value.

According to one embodiment, the dead band output may be selected from a group including a first output mode in which, once the dead band extreme value has been reached, the filtered speed joins the speed defined by the droop, a second output mode in which, once the extreme value of the dead band is reached, the filtered speed may be defined by the droop while maintaining the constant offset of the dead band proportional to the measured speed, and a third output mode in which once the extreme value of the dead band has been reached, the filtered speed joins the speed defined by the droop while following a ramp equivalent to a droop of 2%.

BRIEF DESCRIPTION OF THE DRAWINGS

The other objectives, characteristics, and advantages of the present application will become apparent on reading the following description, given solely by way of non-limiting examples, and made with reference to the accompanying drawings, in which:

FIG. 1 illustrates a flowchart of a method of determining a static response profile of a rotating electrical machine according to an embodiment of the present application;

FIG. 2 illustrates a graph representing a set of functions of a universal speed filter determined according to the method of FIG. 1; and

FIG. 3 shows in detail an example of the application of the universal speed filter of FIG. 2.

DETAILED DESCRIPTION

In the following description, the term “measured speed value Vm” is understood to mean the image of the frequency of the electrical network as seen by the controller, the real value of the rotation of the shaft of the rotating machine. The measured speed value Vm is expressed as a percentage (%) of the speed of the electrical generating unit with respect to the nominal speed of the rotating machine. A value of 100% of speed corresponds to 50 Hz or 60 Hz depending on the country.

The power contribution to be provided by each power generating group may be defined by its own droop, i.e., the ratio between the power variation and the frequency variation of the power grid expressed as a percentage (%). For example, a 4% droop means that a 4% change in the speed of the rotating machine will result in a 100% change in the nominal power of the rotating machine. Thus, an over-speed of the electrical network of 1%, that means 0.5 Hz, will imply a 25% decrease in the nominal power of the rotating machine.

The droop may be adjusted between 2% and 20%. Thus, with a droop of 20% and an over-speed of the electrical network of 1%, that means 0.5 Hz, may imply a 5% decrease in the nominal power of the rotating machine. Similarly, with a 2% droop and an over-speed of the power grid of 1%, that means 0.5 Hz, may imply a 50% reduction in the nominal power of the rotating machine.

FIG. 1 shows a flow chart of a method 10 for determining a static response profile of a rotating electric machine connected to an electrical network capable of responding to variations in the frequency of the electrical network. Hereafter the droop response profile will be called a speed profile or a universal speed filter.

As illustrated in FIG. 1, the control method of the rotating machine may include a first step 12 for recovering a measured speed value Vm and a second step 14 for determining a number of droop response parameters, dependent on the measured speed Vm of the rotating electrical machine.

in step 14, low and high parameters of the droop response corresponding to under-speed and over-speed are determined:

• • the value of low dead band BMB and high dead band BMH, expressed as a percentage (%) of the measured speed value Vm,
  • the value of the median droop SM, low SB and high SH, expressed as a percentage (%) of the measured speed value Vm,
  • the value of the low droop limiter LSB and high LSH, expressed as a percentage (%) of the measured speed value Vm,
  • the dead band output mode for the under-speed and the over-speed, selected from SBM1, SBM2, SBM3, it is also possible to choose a different dead band output mode for under and over speed,
  • the value of the low breaking point PCB and high PCH of the non-linear droop, expressed as a percentage (%) of the measured speed value Vm.

A dead band BM is defined as an inhibition of the power response of the power generation group within a given speed range. Thus, three types of dead bands are defined:

• • A minimal dead band, applied by default, corresponding to the smallest acceptable dead band, for example between +/−0.02% of the measured speed value, that means+/−10 mHz with respect to the nominal frequency. This minimal dead band makes it possible to avoid the load variations of the rotating machine for small variations in the frequency of the electrical network.
  • A variable, symmetrical or asymmetric dead band referenced to the nominal speed, and for example between +/−1% of the measured speed value Vm, that means+/−500 mHz.
  • A fixed, symmetrical or asymmetric dead band, for example between −6% and 1% of the measured speed value Vm, that means between −3 Hz and 0.5 Hz.

The choices of the BM dead band are exclusive, therefore if the variable dead band is activated, then the fixed and default dead bands are disabled. Similarly, when the fixed and variable dead bands are deactivated, the default band BM₁ is activated.

The value of the low dead band BMB is, for example, between 0.02% and 6% of the measured speed value Vm.

The value of the high dead band BMH is, for example, between 0.02% and 1% of the measured speed value Vm.

The values of median droop SM, low droop SB and high droop SH are, for example, between 2% and 20% of the measured speed value Vm.

Droop response limitations makes it possible to limit the contribution of the load from a percentage value of the measured speed Vm to over-speed and/or under-speed by limiting the filtered speed to a constant value. In addition, in the case of over-speed above 101%, the droop response limitation may be deactivated to prevent the rotating machine from operating at high load and speed. Thus, for example, a value of the low droop limiter LSB of between 96% and 100% of the filtered speed value may be selected, and a value of the high droop limiter LSH of between 100% and 104% of the filtered speed value.

The SBM dead band output represents the behavior of the rotating machine at the output of the dead band BM, that is, when the speed value measured in % exceeds the predefined dead band BM.

Thus, three modes of dead band output are defined:

• • The first output mode SBM1, referred to as the step, in which, once the extreme value of the dead band BM has been reached, the filtered speed may have as a value the modulated speed according to the droop applied.
  • The second output mode SBM2, referred to as the rail, in which, once the extreme value of the dead band BM has been reached, the filtered speed may be defined by the droop applied in proportion to the measured speed while maintaining the constant offset of the dead band.
  • The third output mode SBM3, referred to as ramp mode, in which, once the extreme value of the dead band BM has been reached, the filtered speed may have the value of the measured speed modulated by the applied droop while following a ramp equivalent to a droop of 2%.

Thus, in over-speed and under-speed, it is possible to choose identical or different dead band output modes.

The values for the low breaking point PCB and high PCH of the non-linear droop may be selected between 0% and 10% of the measured speed value.

We define a variable droop by default of 4% and adjustable over a range of between 2% and 20% applied over the entire operating range, and a nonlinear droop composed of three speed ranges having their respective droop and delimited by two points of inflection on either side of the nominal speed.

The static response parameters may be determined either by the so-called “TSO” transmission system operator (“TSO”) or by the operator.

Some of the droop response parameters may be set up or changed by the operator and other droop response parameters may be set in the software or controller without being able to be modified.

The method 10 then includes a step 16 for determining the coordinates [X5; Y5] of a point of origin of a graph illustrating a speed profile or response profile in droop, illustrated in FIG. 2. The coordinates [X5; Y5] of the point of origin are defined according to the following equation:

$\begin{array}{cc}\{\begin{array}{c}X5=100\%\\ Y5=100\%\end{array}& \left(\mathrm{Eq}.1\right)\end{array}$

The speed profile, illustrated in FIG. 2, is a graph defined by a set of points of coordinates [Xi; Yi], where “i” is an integer between 0 and 10, the abscissa being the value of the measured speed Vm, in %, corresponding to the image of the frequency of the electrical network, and for ordinates the value of the filtered speed Vf, in %, corresponding to the measured speed Vm modulated by the response parameters in droop.

The term “measured speed value Vm” means the real rotational value of the rotating machine shaft, expressed as a percentage (%) of speed with respect to the nominal speed of the rotating machine which is equivalent to 100%.

The term “filtered speed value Vf” is understood to mean the speed value expressed as a percentage (%) of speed with respect to the nominal speed of the rotating machine modulated by the various statistic response parameters determined in step 14.

As shown in FIG. 2, the speed profile is centered on the coordinates [X5; Y5] of the origin point corresponding to the measured speed Vm nominal of 100%. The corresponding filtered speed Vf is also 100%. Alternatively, the point of origin [X5; Y5] may be adjusted in a range between 99% and 101% of the measured speed Vm.

The method includes calculating the coordinates [X4; Y4] to [X0; Y0] from the first to the fifth point respectively in the case of under-speed and the calculation of the coordinates [X6; Y6] to [X10; Y10] at the first to fifth point respectively in the case of over-speed.

As illustrated in FIG. 1, the method may include steps 18 to 32 for calculating the points of coordinates [X4; Y4] to [X0; Y0] in the case of under-speed and steps 34 to 48 for calculating the points of coordinates [X6; Y6] to [X10; Y10] in the case of over-speed.

In step 18, the coordinates [X4; Y4] from a first under-speed point as a function of the low dead band BMB.

Thus, for example, in the case of under-speed not exceeding the selected low dead band BMB, the value of the filtered speed at point Y4 will correspond to the nominal speed of 100%. The coordinates of the first point [X4; Y4] according to the following equation:

$\begin{array}{cc}\{\begin{array}{c}X4=100\%-\mathrm{BMB}\\ Y4=100\%\end{array}& \left(\mathrm{Eq}.2\right)\end{array}$

Outside the BMB low dead band, the real speed delta corresponds to a filtered delta of speed, that is to say to the delta of measured speed multiplied by a gain of the droop. The gain of the droop may be the ratio between the intrinsic droop of the rotating machine, for example equal to 4%, divided by the desired droop.

Thus, in step 20, the value of the gain of the median droop GSM and the value of the gain of the low droop GSB may be calculated as a function of the median droop SM and the low SB respectively according to the following equations:

$\begin{array}{cc}\mathrm{GSM}=\frac{4\%}{\mathrm{SM}}& \left(\mathrm{Eq}.3\right)\\ \mathrm{GSB}=\frac{4\%}{\mathrm{SB}}& \left(\mathrm{Eq}.4\right)\end{array}$

For example, for a desired low or median droop of between 2% and 20%, the low droop gain GSB and the median droop gain GSM may be between 2 and 0.2 respectively, for example equal to 1, for example equal to 0.5.

In step 22, the coordinates [X3; Y3] of a second point under-speed as a function of the mode of output of the SBM dead band selected in step 14.

If the dead band SBM1 output of step type has been selected in step 14, the coordinates [X3; Y3] of the second point according to the following Equation Eq. 5:

$\begin{array}{cc}\{\begin{array}{c}X3=X4\\ Y3=100-\mathrm{MIN}\left(100-\mathrm{LSB};\mathrm{BMB}·\mathrm{GSM}\right)\end{array}& \left(\mathrm{Eq}.5\right)\end{array}$

The abscissa X3 of the second point may be equal to the abscissa X4 of the first point previously determined in step 18.

The y-coordinate Y3 of the second point may be equal to 100 minus the minimum value between (100 minus the value of the low droop limiter LSB) and the value of the low dead band BMB multiplied by the median GSM droop gain.

If the rail-type SBM2 dead band output has been selected in step 14, the coordinates [X3; Y3] of the second point according to the following Equation Eq. 6:

$\begin{array}{cc}\{\begin{array}{c}X3=X4\\ Y3=Y4\end{array}& \left(\mathrm{Eq}.6\right)\end{array}$

The second coordinate point [X3; Y3] may be coincident with the first point of coordinates [X4; Y4] previously determined in step 18.

If the dead band output SBM3 in ramp mode was selected in step mode 14 and the gain value of the median droop is different from 2, then we have the coordinates of the second point [X3; Y3] according to the following Equation Eq. 7:

$\begin{array}{cc}\{\begin{array}{c}X3=100-\mathrm{MIN}\left(\frac{100-\mathrm{LSB}}{2};\frac{\mathrm{BMB}·\mathrm{GSM}}{2-\mathrm{GSM}}\right)\\ Y3=100-\left(2·\left(X4-X3\right)\right)\end{array}& \left(\mathrm{Eq}.7\right)\end{array}$

If the value of gain of the median droop is equal to 2, the low dead band cannot be ramped, thus we may retake the coordinates [X3; Y3] of the second point defined in Equation Eq. 6.

In step 24, when the output of dead band SBM1 of step type has been selected, the value of the low breaking point PCB may be compared with the value of the low dead band BMB.

If the value of the low breaking point PCB is lower than the value of the low dead band BMB, the we recalculate the coordinates [X3; Y3] of the second point per the following Equation Eq. 8:

$\begin{array}{cc}\{\begin{array}{c}X3=X4\\ Y3=100-\mathrm{MIN}\left(100-\mathrm{LSB};\mathrm{BMB}·\mathrm{GSB}\right)\end{array}& \left(\mathrm{Eq}.8\right)\end{array}$

In step 26, we recalculate the coordinates [X2; Y2] of a third point in under-speed, corresponding to the low breaking point of the non-linear droop, according to the following Equation Eq. 9:

$\begin{array}{cc}\{\begin{array}{c}X2=\mathrm{MAX}\left(100-\mathrm{PCB};X3-\frac{Y3-\mathrm{LSB}}{\mathrm{GSM}}\right)\\ Y2=Y3-\left(X3-X2\right)·\mathrm{GSM}\end{array}& \left(\mathrm{Eq}.9\right)\end{array}$

In step 28, the value of the abscissa X3 of the second point is compared with (100−PCB).

If the value 100−PCB is greater than the value of the abscissa X3 of the second point, we recalculate the coordinates [X2; Y2] per the following Equation Eq. 10:

$\begin{array}{cc}\{\begin{array}{c}X2=X3\\ Y2=Y3\end{array}& \left(\mathrm{Eq}.10\right)\end{array}$

In step 30, the coordinates [X1; Y1] of a fourth under-speed point may be calculated, corresponding to the under-speed droop limiter, per the following Equation Eq. 11:

$\begin{array}{cc}\{\begin{array}{c}X1=\mathrm{MAX}\left(90;X2-\frac{Y2-\mathrm{LSB}}{\mathrm{GSB}}\right)\\ Y1=Y2-\left(X2-X1\right)·\mathrm{GSB}\end{array}& \left(\mathrm{Eq}.11\right)\end{array}$

In step 32, we recalculate the coordinates [X0; Y0] of a fifth point in under-speed, corresponding to the under-speed limit point of the filter, per the following Equation Eq. 12:

$\begin{array}{cc}\{\begin{array}{c}X0=90\\ Y0=\mathrm{MAX}\left(\mathrm{LSB};Y1\right)\end{array}& \left(\mathrm{Eq}.12\right)\end{array}$

Thus, each segment defined by two points corresponds to a function modulated by the functions that precedes it.

The steps 34 to 48 represent the steps of calculating the points of coordinates [X6; Y6] to [X10; Y10] in the case of over-speed.

In step 34, the coordinates [X6; Y6] of a first point in over-speed, as a function of the high dead band BMH.

Thus, for example, in case of over-speed not exceeding the selected high BMH dead band, the value of the speed filtered at point Y6 may correspond to the nominal speed of 100%. The coordinates [X6; Y6] of the first point per the following Equation:

$\begin{array}{cc}\{\begin{array}{c}X6=100\%+\mathrm{BMH}\\ Y6=100\%\end{array}& \left(\mathrm{Eq}.13\right)\end{array}$

Outside the selected BMH high dead band, the real speed delta corresponds to a filtered delta of speed, that is to say to the delta of measured speed multiplied by a gain of the droop. The gain of the droop may be the ratio between the intrinsic droop of the rotating machine, for example equal to 4%, divided by the desired droop.

Thus, the median GSM droop gain value, calculated in step 20, may be applied.

In step 36, the value of the gain of the high droop GSH as a function of the high droop SH may be calculated per the following Equation:

$\begin{array}{cc}\mathrm{GSH}=\frac{4\%}{\mathrm{SH}}& \left(\mathrm{Eq}.14\right)\end{array}$

For example, for a desired high droop between 2% and 20%, the high droop gain GSH may be between 2 and 0.2.

In step 38, the coordinates [X7; Y7] of a second point, in over-speed, as a function of the output mode of the dead band SBM may be determined in step 14.

If the step-mode dead band output SBM1 has been selected in step 14, the coordinates [X7; Y7] of the second point may be calculated per the following Equation Eq. 15:

$\begin{array}{cc}\{\begin{array}{c}X7=X6\\ Y7=100+\mathrm{MIN}\left(\mathrm{LSH}-100;\mathrm{BMH}·\mathrm{GSM}\right)\end{array}& \left(\mathrm{Eq}.15\right)\end{array}$

The abscissa X7 of the second point may be equal to the abscissa X6 of the first point previously determined in step 34.

The y-coordinate Y7 of the second point may be equal to 100 plus the minimum value between (the value of the high-droop limiter LSH minus 100) and (the value of the high dead band BMH multiplied by the median GSM droop gain calculated at 1 Step 20).

If the rail-mode SBM2 dead band output has been selected in step 14, the coordinates [X7; Y7] of the second point may be calculated per the following Equation Eq. 16:

$\begin{array}{cc}\{\begin{array}{c}X7=X6\\ Y7=Y6\end{array}& \left(\mathrm{Eq}.16\right)\end{array}$

The second coordinate point [X7; Y7] may be coincident with the first point of coordinates [X6; Y6] previously determined in step 34.

If the ramp mode SBM3 output of dead band has been selected in step 14 and the value of the gain of the median droop is different from 2, the coordinates [X7; Y7] of the second point may be calculated per the following Equation Eq. 17:

$\begin{array}{cc}\{\begin{array}{c}X7=100+\mathrm{MIN}\left(\frac{\mathrm{LSH}-100}{2};\frac{\mathrm{BMH}·\mathrm{GSM}}{2-\mathrm{GSM}}\right)\\ Y7=100+\left(2·\left(X7-X6\right)\right)\end{array}& \left(\mathrm{Eq}.17\right)\end{array}$

If the value of the gain of the median droop is equal to 2, the high dead band BMH cannot be caught, we retake the coordinates [X7; Y7] of the second point defined in the Equation Eq. 16.

In step 40, when the step mode dead band output SBM1 has been selected, the value of the high breaking point PCH may be compared with the value of the high dead band BMH.

If the value of the high breaking point PCH is less than the value of the high dead band BMH, the coordinates [X7; Y7] of the second point may be calculated per the following Equation Eq. 18:

$\begin{array}{cc}\{\begin{array}{c}X7=X6\\ Y7=100+\mathrm{MIN}\left(\mathrm{LSH};\mathrm{BMH}·\mathrm{GSH}\right)\end{array}& \left(\mathrm{Eq}.18\right)\end{array}$

In step 42, the coordinates [X8; Y8] of a third point may be calculated, in over-speed, corresponding to the high breaking point of the non-linear droop, per the following Equation Eq. 19:

$\begin{array}{cc}\{\begin{array}{c}X8=\mathrm{MIN}\left(100+\mathrm{PCH};X7+\frac{\mathrm{LSH}-Y7}{\mathrm{GSM}}\right)\\ Y8=Y7+\left(X8-X7\right)·\mathrm{GSM}\end{array}& \left(\mathrm{Eq}.19\right)\end{array}$

In step 44, the value of the abscissa X7 of the second point may be compared with (100+PCH).

If the value 100+PCH is less than the value of the abscissa X7, the coordinates [X8; Y8] of the third point may be calculated per the following Equation Eq. 20:

$\begin{array}{cc}\{\begin{array}{c}X8=X7\\ Y8=Y8\end{array}& \left(\mathrm{Eq}.20\right)\end{array}$

In step 46, the coordinates [X9; Y9] of a fourth point may be calculated, corresponding to the over-speed droop limiter, per the following Equation Eq. 21:

$\begin{array}{cc}\{\begin{array}{c}X9=\mathrm{MIN}\left(110;X8-\frac{\mathrm{LSH}-Y8}{\mathrm{GSH}}\right)\\ Y9=Y8+\left(X9-X8\right)·\mathrm{GSH}\end{array}& \left(\mathrm{Eq}.21\right)\end{array}$

In step 48, the coordinates [X10; Y10], of a fifth point may be calculated, corresponding to the over-speed limit point of the filter, per the following equation Eq. 22:

$\begin{array}{cc}\{\begin{array}{c}X10=110\\ Y10=\mathrm{MIN}\left(\mathrm{LSH};Y9\right)\end{array}& \left(\mathrm{Eq}.22\right)\end{array}$

Thus, each segment defined by two points corresponds to a function modulated by the functions that precedes it.

As illustrated in FIG. 2, the output of the dead band beyond the point of coordinates [X4; Y4] in the case of under-speed and beyond the point of coordinates [X6; Y6] in case of over-speed, has three possible functions. These different modes allow one to obtain distinct speed profiles. FIG. 2 shows, in solid lines, the so-called ramp mode of the output of the dead band SBM3, in dotted lines, the so-called step mode of output of the dead band SBM1 and in bold dashed lines, the rail mode of the dead band SBM2.

In the rail mode, the coordinate points [X7; Y7] of the dead band may be coincident with the coordinate point [X6; Y6] defining the dead band for over-speed. Similarly, for the under-speed, the point of exit of the dead band of coordinates [X3; Y3] may be coincident with the point of coordinates [X4; Y4] defining the dead band. Thus, outside the dead band, a filtered delta of speed corresponds to a delta of measured speed multiplied by the gain of the droop.

In the ramp mode, the output of the dead band may be set to over-speed by the segment of coordinates [X6; Y6] and [X7; Y7] corresponding to the segment between the first and second point or under-speed by the segment of coordinates [X4; Y4] and [X3; Y3] corresponding to the segment between the first and second point.

The filtered speed joins the real speed modulated by the droop gain following a ramp equivalent to a 2% droop.

In the step mode, the output of the dead band may be set to over-speed by the segment of coordinates [X6; Y6] and [X7; Y7] corresponding to the segment between the first and the second point, or under-speed by the segment of coordinates [X4; Y4] and [X3; Y3] corresponding to the segment between the first and second point. The filtered speed joins the real speed modulated by the droop gain along a step from the coordinate point [X6; Y6] in over-speed or from the point of coordinates [X4; Y4] at under-speed.

Thus, the two segments defined by the coordinates [X7; Y7], [X8; Y8] of the second and third points and by the coordinates [X8; Y8], [X9; Y9] of the third and fourth points for the over-speed, or the two segments defined respectively by the coordinates [X3; Y3], [X2; Y2] of the second and third points and by the coordinates [X2; Y2], [X1; Y1] of the third and fourth points for the under-speed, combine two functions related to the droop, namely:

• • Nonlinear droop: between the points of coordinates [X7; Y7] and [X8; Y8], the median droop gain may be applied and between the points of coordinates [X8; Y8] and [X9; Y9] the high droop gain may be applied. The point of coordinates [X8; Y9] corresponds to the breaking point between the two droop segments.
  • Variable droop: in this case, the median and high droop gains may be equal, creating a single segment between the points of coordinates [X7; Y7] and [X9; Y9] in case of over-speed or between points of coordinates [X3; Y3] and [X1; Y1] at under-speed.

The segment defined by the points of coordinates [X9; Y9] and [X10; Y10] represents the high droop limiter, in which zone the filtered speed may be constant regardless of the real measured speed variation.

The segment defined by the points of coordinates [X1; Y1] and [X0; Y0] represents the low droop limiter, in which zone the filtered speed is constant regardless of the real measured speed variation.

The graph illustrated in FIG. 2 represents the set of functions of a universal speed filter obtained by the method described with reference to FIG. 1.

FIG. 3 illustrates a particular case of the universal filter of FIG. 2, in which a mode of output of the dead band in the ramp mode has been selected. The set of points of coordinates [X4; Y4] to [X0; Y0] in the case of under-speed and coordinate points [X6; Y6] to [X10; Y10] in the case of over-speed may be calculated according to the steps 18 to 48 previously described.

As soon as the determination method illustrated in FIG. 1 has elaborated the response profile in the form of a droop or universal speed profile illustrated in FIG. 2, it may be displayed on a man-machine interface (HMI). It would also be possible to display on this man-machine interface a theoretical power response profile corresponding to the universal speed profile displayed. It will be noted that the coordinates of the points of this power response profile may be obtained from the coordinates of the points of the universal speed profile and based on the relationship between the filtered speed variation and the power variation inherent in the definition of the droop.

In the method herein, it would be possible to provide a limitation to a minimum power defined by the operator using the droop limiter. The difference between the real power and the minimum power then may be converted into a permissible variation to define this limitation.

In general, the method herein makes it possible to integrate a number of functions related to the measured speed, such as in particular the value of the dead band and the value of the droop. From this, the method may determine a universal speed profile, also known as a universal droop response profile or universal speed filter. This universal speed profile according to the method herein is thus obtained in various ways: either all of the parameters of the frequency response are predefined, for example specified in the transport network manager or determined by the operator. In some cases, a few or none of these parameters of the frequency response may be unspecified, then the universal speed profile may be developed using default parameters, for example a default dead band of 10 mHz, with rail mode band dead output and/or a droop equal to 4%, or the parameters are defined according the two preceding ways.

According to the described method, if different parameters are selected in over-speed and under-speed, the determined speed profile may be asymmetrical around the coordinate origin point [X5; Y5]. It thus may be possible to obtain a different behavior from the rotating electrical machine in over-speed and under-speed. Asymmetry may be particularly attractive for markets where over-speed and under-speed responses represent different products and services.

The method herein thus makes it possible to calculate automatically and independently the points of coordinates [X0; Y0] through [X4; Y4] at under-speed with respect to the nominal speed and the points of coordinates [X6; Y6] to [X10; Y10] at over-speed with respect to nominal speed.

The independence of the calculation makes it possible to obtain the asymmetry of the droop profile.

In addition, the simultaneous calculation of the coordinates of the points makes it easy to integrate the modifications of the response parameters into droop. When a parameter is changed, for example in the electrical network manager, the method thus may recalculate the set of coordinates of the points defining the universal speed profile, which makes the method herein particularly flexible. By virtue of the method herein, when at least one of the droop response parameters evolves, the method readjusts or modifies this parameter and automatically recalculates the set of coordinates of the points defining the universal speed filter.

It should be apparent that the foregoing relates only to certain embodiments of the present application and the resultant patent. Numerous changes and modifications may be made herein by one of ordinary skill in the art without departing from the general spirit and scope of the invention as defined by the following claims and the equivalents thereof.

Claims

1. A method for determining a droop response profile of a rotating machine supplying electricity to an electrical network having a network frequency varying on either side of a nominal frequency, wherein a measured speed value (Vm) of the rotation of the rotating machine corresponding to the frequency of the electrical network and wherein response parameters to the measured speed value are defined, characterized in that the profile is a graph centered on a coordinate point of origin ([X5; Y5]) between 99% and 101% of the measured speed value and defined by at least two coordinate points ([X4; Y4], [X0; Y0]) in case of under-speed and by at least two coordinate points ([X6; Y6], [X10; Y10]) in case of over-speed, each point having a speed value as a percentage of the measured speed value, and for ordinates a speed value filtered as a percentage of the measured speed value modulated by at least one of the response parameters, the response parameters comprising at least a high dead band (BMH) value and a low dead band (BMB) value on either side of the point of origin ([X5;Y5]), a low droop, a median droop, and a high droop (SB, SM, SH) of the rotating machine, a low and a high droop limit value (LSB, LSH), at least one dead band output mode (SBM1, SBM2, SBM3), and a low and a high breaking point (PCB, PCH) of a nonlinear droop.

2. The method of claim 1, wherein the coordinates ([X4; Y4]) of a first point in the case of under-speed correspond to the low dead band value and have an abscissa (X4) equal to subtraction of 100% of the measured speed value with the low dead band value, and for ordinates (Y4) equal to 100% of the measured speed value.

3. The method of claim 2, wherein a median droop gain (GSM) value and a low droop gain (GSB) value correspond to a ratio between the intrinsic droop of the rotating machine and, respectively, the median droop (SM) and the low droop (SB).

4. The method of claim 3, wherein coordinates ([X3; Y3]) of a second point in the case of under-speed correspond to the output of the dead band as a function of the at least one dead band output mode (SBM), the low limit droop (LSB) value, the low dead band (BMB) value, the median droop gain (GSM) value, and the low breaking point (PCB) value.

5. The method of claim 4, wherein coordinates ([X2; Y2]) of a third point in the case of under-speed correspond to the low breaking point of the nonlinear droop as a function of the coordinates of the second point ([X3;Y3)], the low breaking point (PCB) value, the low droop limit (LSB) value, and the median droop gain (GSM) value.

6. The method of claim 5, wherein coordinates ([X1; Y1]) of a fourth point in the case of under-speed correspond to the low droop limit value as a function of the coordinates of the third point ([X2; Y2]) and the low droop gain (GSB) value.

7. The method of claim 6, wherein coordinates ([X0; Y0]) of a fifth point in the case of under-speed correspond to a limit point of the response profile as a function of the coordinates of the fourth point ([X1; Y1]) and the low droop limit (LSB) value.

8. The method of claim 7, wherein coordinates ([X6; Y6]) of a first point in the case of an over-speed correspond to the high dead band value and have the abscissa (X6) equal to the addition of 100% of the measured speed value to the high dead band (BMH) value, and for ordinate (Y6) equal to 100% of the measured speed value.

9. The method of claim 8, wherein the high droop gain (GSH) value corresponds to a ratio between the intrinsic droop of the rotating machine and the high droop (SH).

10. The method of claim 9, wherein coordinates ([X7; Y7]) of a second over-speed point correspond to the output of the dead band as a function of the dead band output mode (SBM), the high droop limit value (LSH), the high dead band (BMH) value, the high droop gain (GSH) value, and the high breaking point (PCH) value.

11. The method of claim 10, wherein coordinates ([X8; Y8]) of a third over-speed point correspond to the high breaking point of the non-linear droop as a function of the coordinates of the second point ([X7; Y7]) of the high breaking point (PCH) value, the high droop limit (LSH) value, and the median droop gain (GSM) value.

12. The method of claim 11, wherein coordinates ([X9; Y9]) of a fourth over-speed point correspond to the high droop limit value as a function of the coordinates of the third point ([X8; Y8]), the high droop limit (LSH) value, and the high droop gain (GSH) value.

13. The method of claim 12, wherein coordinates ([X10; Y10]) of a fifth point in over-speed correspond to a limit point of the response profile as a function of the coordinates of the fourth point ([X9; Y9] and the high droop limit (LSH) value.

14. The method of claim 1, wherein the low dead band (BMB) value is between 0.02% and 6% of the measured speed value (Vm).

15. The method of claim 1, wherein the high dead band (BMH) value is between 0.02% and 1% of the measured speed value (Vm).

16. The method of claim 1, wherein the median droop (SM) value, the low droop (SB) value, and the high droop (SH) value are between 2% and 20% of the measured speed value (Vm).

17. The method of claim 1, wherein the low breaking point (PCB) value and the high breaking point (PCH) value of the non-linear droop are selected from 0% to 10% of the measured speed value.

18. The method of claim 1, wherein the low droop limit (LSB) value is between 96% and 100% of the filtered speed value.

19. The method of claim 1, wherein the high droop limit (LSH) value is between 100% and 104% of the filtered speed value.

20. The method of claim 1, wherein the at least one dead band output mode (SBM) is selected from the group comprising a first output mode (SBM1) in which, once a dead band extreme value has been reached, the filtered speed reaches the speed defined by the droop, a second output mode (SBM2), in which, once the deal band extreme value has been reached, the filtered speed is defined by the droop while maintaining a constant offset of the dead band proportional to the measured speed value, and a third output mode (SBM3), in which, once the extreme value of the dead band has been reached, the filtered speed joins the speed defined by the droop while following a ramp equivalent to a 2% droop.