GB2302964A

GB2302964A - Regulating turbine fuel supply

Info

Publication number: GB2302964A
Application number: GB9613689A
Authority: GB
Inventors: Peter Wittmann
Original assignee: MTU Motoren und Turbinen Union Muenchen GmbH
Current assignee: MTU Aero Engines GmbH
Priority date: 1995-06-30
Filing date: 1996-06-28
Publication date: 1997-02-05
Anticipated expiration: 2016-06-28
Also published as: DE19523845A1; ITMI961217A1; IT1284018B1; FR2736093B1; GB9613689D0; GB2302964B; FR2736093A1; ITMI961217A0; DE19523845C2

Description

2302964 FUEL SUPPLY REGULATION The invention relates to a process and an

apparatus for regulating the supply of fuel for engines in acceleration and deceleration procedures, in which predetermined maximum values for turbine rotational speeds and temperatures are set with different flight attitudes, the differences of these from the actual values being fed to a PID regulator which controls a fuel supply valve in a fuel supply line to the burner of the engine, in particular a turbine.

A system of this type is known from DE-PS 38 30 804. The aim of the process disclosed here is to avoid forbidden temperature overshoots during acceleration of an engine, using for this purpose a PID regulator which has as its input variables essentially weighted or modulated deviations from a set-point or desired value and differentiated actual values of rotational speed and temperature. The desired value difference for temperature is the difference between the measured temperature in the high-pressure turbine and a maximum permitted temperature and for rotational speed it is the difference between the measured rotational speed of the high-pressure turbine and a maximum permitted rotational speed. The maximum permitted temperature is dependent on the flight states of the aircraft or the operational states of the engine. Moreover the measured actual value is pyrometrically determined.

A disadvantage of this process is that it does not take into account the differences in heating and cooling behaviour in the housing and the blades during acceleration and deceleration of the engine. These differences mean that the determining of the actual value of the temperature at the blades in an engine is regularly subject to errors; generally the temperature measured will be too low during an acceleration phase, which has the result that the turbine input stage is thermally overloaded and the permitted maximum temperature value is exceeded. Therefore, not only is the service life of the turbine stage disadvantageously reduced but during acceleration a rotational speed overshoot is generated, resulting in a thrust overshoot which hinders controlled flight manoeuvres. With the conventional regulation system the thermal state of the engine housing is not taken into consideration in the case of deceleration, so that with renewed acceleration the initial thermal state of the housing is assumed to be unchanged, which leads to considerable mismatchings of the fuel supply in the renewed acceleration phase and may thermally overload the turbine blades.

It is the aim of the invention to provide a process and an apparatus for controlling the supply of fuel for engines, where heating or coolingdependent errors in the determination of actual values of the temperatures in an engine are taken into account, in order to reduce thrust or temperature overshoots, resulting in an improved service life of the engine while still making full use of the maximum permitted thermal loading and while still guaranteeing reliable flight and rolling manoeuvres.

According to the invention, the maximum permitted temperature T, of the high pressure turbine is modulated using the heating or cooling behaviour of the engine housing, either measured or simulated, and the resultant temperature and rotational speed differences are supplied to the controller, e.g. a PID regulator.

The advantage of this solution is that the regulation behaviour is not determined by idealised and abstract temperature differences in the rotor, in particular in the high-pressure turbine, but the is control includes the actually existing heating or cooling behaviour of the engine housing. The effect of the heat radiation from the engine housing on the temperature detection of the turbine input stage is slight at low initial rotational speed and on acceleration increases more slowly than the temperature increase of the turbine stage. The heat radiation values which form the basis of the temperature detection, and hence the actual values of the temperature, are made up of the heat radiation component from the input stage of the high-pressure turbine and the scattered values from the housing, so that the temperature detection in the acceleration phase measures actual values which are too low and hence outputs a higher value of the temperature difference from the maximum permitted turbine input temperature than is really present. The same is true with reverse signs for the deceleration phase, although this does not have any direct negative influence on the quality of control in the case of deceleration operations as this control is not determined analogously by a minimum temperature limit value. On the other hand the measures taken ensure that in the case of decelerations the varying thermal state of the turbine housing is advantageously tracked at all times and hence with renewed acceleration the starting point matches the thermal behaviour of the engine housing. With the present invention therefore the control procedure is more closely matched to actual engine conditions.

In a preferred implementation of the process for weighting the maximum permitted temperature a temperature reduction (AT) is delayed by means of a filter, preferably of the first order, the filter reproducing the heat behaviour of the engine housing in the case of heat absorption of the engine housing in a heating or acceleration phase or in the case of heat emission from the engine housing in a cooling or deceleration phase with different time constants (T,, T.,) as between different operational states of the engine with different roll or flight manoeuvres of an aircraft. This has the advantage that the heat radiation of the engine housing and its effect on the measured actual value does not need to be constantly detected by appropriate additional measuring devices and with corresponding expenditure on measurement, but rather with fixed time constants, determined once, the effect of the engine housing is taken into account.

These time constants are preferably determined and stored in test runs of the engine for all conceivable transitions from one engine state to the other and automatically called up when required. This has the advantage that in flight the engine does not have to carry any additional, heavy measuring devices for detecting an increase or decrease in the housing temperature, but merely a small semi-conductor chip which is light compared with a measuring device, from which the previously determined time constants can be recalled.

In a simple version of the control process of the invention preferably two time constants are determined and stored in dependence on the difference between the actual value (TI) of the temperature and the desired or maximum value (Tm), namely a time constant (T.) for heating of the engine housing during acceleration and a time constant (Ta) for cooling of the engine housing during deceleration. A heating of the engine housing means a direct supply of energy until the actual value (TI) of the temperature has reached the maximum value (Tm) of the temperature except for a threshold value (TK). For this purpose a test rig is used first of all and then in the case of engine tests a fine adjustment is of the time constants T,, and T. is carried out by means of the engine behaviour.

In another preferred embodiment of the process of the invention the magnitude of the temperature reduction (AT) has an upper limit (ATmax) and is proportional to the difference (ND) between the actual value (NI) of the rotational speed and a rotational speed threshold value (NscH) which is smaller than the maximum permitted rotational speed (Nm) by 5 to 20%. The temperature reduction is therefore dependent on the distance of the high-pressure rotational speed from a constant threshold value. This threshold value has the advantageous effect that after reaching the rotational speed threshold value the weighting effect of the speed on the maximum permitted temperature is set at 0 and hence the weighting is no longer dependent on the rotational speed. Above this threshold value no reduction in the maximum permitted temperature is required for the control behaviour of the PID regulator. The high-pressure turbine rotational speed, which is the same as the high-pressure compressor speed, is preferably used because this rotational speed is the signal with the dynamically fastest reaction to a change in the fuel supply. The difference between the high-pressure turbine rotational speed and the constant threshold value determines the starting value for the weighted maximum permitted temperature on acceleration and thus represents a gauge for the initial heating state of the engine.

The temperature reduction (AT) is also preferably weighted by a factor which is proportional to the difference between the maximum allowed value of the temperature (Tm) and the measured actual value of the temperature (TI) and assumes normalised values between 0 and 1 between a smallest temperature reduction (ATmin) and a largest temperature reduction (ATinax). This has is the advantage that the weighting dependent on the rotational speed difference can be modulated using the normalised values between 0 and 1 for the temperature difference. A value is selected as the largest temperature reduction (AT,,a,) caused by the temperature difference that is equal to the upper limit (AT..) of the temperature reduction caused by the rotational speed difference, so that advantageously in the case of adulteration or errors in the input values for the rotational speed difference, such as erroneous threshold value adjustments or errors in actual value detection of the rotational speed, the temperaturedependent logic unit, with the corresponding proportional element ensures that the temperature reduction in each case goes to zero and the maximum temperature loading for the turbine stage is completely made use of without overshooting.

In a further preferred embodiment of the process of the invention the increase in temperature reduction (AT) due to the rotational speed difference (ND) together with the normalised values resulting from the temperature difference (T,,) are multiplied at a multiplication point (M) and fed to the filter whose output signal (AT) is subtracted at a first summation point (S,) from the maximum value of the temperature (T,J to form the weighted temperature (Tgew) - This multiplication ensures that the proportional increase between the smallest temperature reduction (ATj, ) and the largest temperature reduction (AT,,a.) can be superimposed on the increase in the temperature reduction due to the rotational speed difference. This also means that when the smallest temperature reduction (ATmi,J is reached and hence when the actual temperature (TI) approaches the maximum permitted temperature (Tm) the temperature reduction (AT) is set to 0 by the above multiplication and the PID regulator outputs the maximum permitted temperature.

According to the invention the process is carried out with a control arrangement as defined by the features of claim 11. For this purpose the engine has a temperature-measuring device, preferably a pyrometer, to determine the actual value of the run-in temperature of the turbine and a speed gauge to determine the actual value of the rotational speed of the turbine. The temperature gauge and the rotational speed gauge measure the actual values in the high-pressure turbine or compressor. A first summation point (adder) (S,) reduces a flight-independent, fixed maximum temperature (T,,) by a temperature reduction (AT) for modulating the maximum value before it is supplied to a PID regulator. The output of the PID regulator controls an adjustment element for the supply of fuel to the engine. A firstorder filter for delaying the temperature reduction is preferably connected upstream of the first summation point (S,), the delay being determined by a heating or cooling time constant (Te, T.) used in accordance with the position of an adjustment member connected upstream of the filter.

This control arrangement has the advantage that it requires no additional components and hence no additional weight with respect to current regulation procedures and arrangements; it merely uses available computer and control capacities and yet operates with substantially improved reality simulation so that the permitted thermal loading of an engine can be completely made use of with full service life of the thermally highestloaded components and the engine allows the greatest possible manoeuvrability for the aircraft.

In a preferred embodiment of the invention a proportional element is connected downstream of a second summation point (S2) which forms the rotational speed difference (NI,) between a threshold value of the rotational speed (NscH) and the measured actual value of the rotational speed (N,), the output of this proportional element being supplied to a multiplication point (M). At a third summation point (S3) the difference (TD) between the maximum value of the temperature (T.) and the actual value of the temperature (TI) is formed. This temperature difference is supplied to a proportional element with a limitation of the value of the minimum and maximum temperature difference. The output of the proportional element is coupled to the multiplication point (M). The output of the multiplication point (M) is connected to the firstorder filter.

This preferred embodiment has the advantage that the reduction applied to the maximum permitted temperature is set to zero before a maximum rotational speed is reached, namely when the threshold value is reached and before the permitted maximum temperature is reached, so that in the final phase of an acceleration operation, for instance the full maximum permitted temperature represents the determining control variable for the PID regulator, and a correspondingly high maximum thrust can be set up with maximum rotational speed without overloading the engine by overshooting the temperature or preventing correct flight manoeuvres by overshooting the rotational speed.

For a better understanding of the invention an embodiment will now be described, referring to the attached drawings, in which:

Fig. 1 shows a control arrangement for carrying out a control procedure in accordance with the present invention; Fig. 2 shows a diagrammatic sketch of an engine with measurement and control signal lines for controlling the fuel supply using the control is procedure; and Fig. 3 shows a time diagram for the variation of high pressure turbine rotational speed and temperature using the control procedure.

Fig. 1 shows a control arrangement 1 for carrying out the control procedure of this example, namely for controlling the supply of fuel for engines during acceleration and deceleration procedures by means of predetermined maximum values for turbine rotational speeds and temperatures which differ according to the instantaneous flight attitude or rolling manoeuvre. The differences between actual values and the set maximum values T, Nm are fed to a PID regulator 10. By means of the supply line 15 the PID regulator 10 controls a fuel supply valve 2, which is shown in Fig. 2, and determines the fuel supply to the burner 14. For this purpose the fuel supply valve is arranged in a fuel supply line 3.

The actual or measured value of the temperature TI is here the temperature at the inlet of the highpressure turbine, which is pyrometrically determined in this example at the blades of the guide baffle of the high-pressure turbine, as shown at 16 in Fig. 2. The actual value of the rotational speed NI is the rotational speed of the high- pressure turbine which is measured in the high-pressure compressor 20, as shown in Fig. 2, position 11. The permitted maximum values T, of the temperature for the different operational states of the engine 17 are weighted or modulated for the different rolling or flying manoeuvres of the aircraft before input into the PID regulator 10, the maximum values Tm being lowered by a temperature reduction AT at a first summation point S, This temperature reduction AT is varied by means of several processing stages or circuits 6, 7 and 8, between a maximum value AT,,.. and zero, depending respectively on a rotational speed difference, a temperature difference and the heating or cooling behaviour of the engine housing 12.

In the embodiment of Fig. 1 first of all threshold values NscH are determined for the rotational speed in different operational states of the engine which are 5 to 20% below the maximum high-pressure turbine rotational speed Nm, i. e.

NscH = Nm - (5 to 200-5) Nm.

At a second summation point S2 the rotational difference ND between the threshold value NscH and the actual value NI of the rotational speed is formed with ND = NscH - NI and fed to a logic circuit with a proportional element forming the processing stage 6. If the actual value NI is above this threshold value NscH then N. becomes negative and the stage 6 sets the temperature reduction AT to zero; hence, irrespective of the rotational speed, the engine can be accelerated to the fixed maximum temperature value Tm. If the actual value NI is below this threshold value NscH the temperature reduction AT increases proportionally to the rotational speed difference ND until it reaches a maximum temperature reduction AT,,,, which in this example is 1000C. The increase in this example is 40C per 1% rotational speed variation. The maximum temperature reduction AT,,, of, for instance, 1000C is therefore reached in this embodiment with a rotational speed at or below 25% of the threshold value Nscm.

The output signal of the logic circuit with the proportional element 6 is fed to a multiplication point M which multiplies the temperature reduction determined by the logic circuit 6 in dependence on the rotational speed a normalised value between 0 and 1 to produce an output T,. These normalised values are dependent on the temperature difference T. between the actual value TI and the permitted maximum value Tm of the high-pressure turbine temperature with TD = T, - TI. This temperature difference TD 'S formed at a third summation point S3. The normalised values between 0 and 1 are calculated from the temperature difference TD in a logic circuit with proportional element 7. As long as the temperature difference TD exceeds the value AT, a,, here 1000C, the output of the logic circuit 7 is set at 1 so that multiplication at the multiplication point M produces a temperature reduction which is in effect determined by means of the proportional element 6 alone on the basis of the rotational speed difference. If on the other hand the temperature difference TD falls below a minimum value AT,in, here 500C, the output of the logic circuit 7 is set to 0 so that multiplication at the multiplication point M produces in effect a zero temperature reduction; in other words it sets the weighted temperature Tgew at the input of the PID regulator to Tm so that the maximum value of the temperature at the blades of the turbine input stage can be achieved without thermal overshooting. In between AT,j, and ATa, the stage 7 produces a linearly rising output between 0 and 1 for multiplication by the rotation-dependent output from the stage 6.

Between the multiplication point M and the summation point S, there is arranged a f ilter 4 of the first order which delays the temperature reduction AT as regards time in order to simulate the heating or cooling behaviour of the engine housing. The filter time constant T is T,, for the heating and T,, for the cooling phases. They are derived from a memory 9 by means of an adjustment element 5 with the switch positions 0 for cooling and 1 for heating and sent to the filter 4, though a software implementation is also conceivable. The decision between the heating and is cooling time constants is dependent on a logic function 8 whose input is connected to the summation point S3 SO that the temperature difference TD forms the input signal. As long as a temperature difference threshold value T. is not exceeded, i.e. the turbine input temperature is close to the maximum temperature, the engine housing is hot, and the output of the logic function 8 is set to 1 so that the heating time constant r, of the engine housing determines the time constant T of the filter 4. If the temperature difference T,, exceeds the temperature difference threshold value T,, then the turbine input stage is at low temperature and the housing emits heat to its surroundings with a cooling time constant T,. In this case the logic function 8 sets its output to 0 and hence the adjustment member 5 into position 0 so that for the time constant T of the filter 4 the cooling time constant Ta of the housing is effective.

The temperature difference threshold value TK 'S preferably set to be the same as the maximum temperature reduction AT,,,ax which consequently means that every decrease in the temperature reduction by multiplication with the temperature-differencedependent evaluation factor takes place in the regime of the heating time constant.

Fig. 2 shows a diagrammatic sketch of an engine 17 with measurement signal lines 18, 19 and a control signal line 15 for control of the fuel supply using the control procedure of the invention. For this purpose measurement signals of the actual values of the rotational speed NI and the temperature T, and maximum values of the temperature T, and threshold values of the rotational speed NscH are fed to a control apparatus 1, as shown in detail in Fig. 1. The output signal of the control arrangement 1 is conveyed via the signal line 15 to an adjustment valve 2 which is arranged in the fuel supply 3 for the burner 14 of the engine 17. The run-in temperature TI of the high-pressure turbine is measured at a blade 16 of the guide baffle of the first turbine stage. The rotational speed NI of the highpressure turbine is determined in the region of the high-pressure compressor 20 at the measuring point 11. Fig. 3 shows a time diagram for the variation of the rotational speed NI of the high-pressure turbine and the temperature TI of the high-pressure turbine using the invention. At time t. a thermal equilibrium state with a rotational speed NI,, for instance an idling rotational speed, will have been set up in the engine. At time t, a new rotational speed value N12 is set to which the engine is to be accelerated. Connected with this specified rotational speed is a maximum value Tm for the temperature of the blades in the high-pressure turbine which must not be exceeded. Since the engine housing is still relatively cold, the actual value of the blade temperature measured in the initial phase is too low, owing to the scattering (or absorbing) effect of the engine housing, and consequently with conventional control too large a temperature difference between the actual temperature value TI and the maximum value Tm is given to a PID regulator which leads disadvantageously to the excess temperature curve TO (shown in dot-dash line) and damages the blades. At the same time an accompanying rotational speed overrun NO prevents the carrying out of correct flight manoeuvres. In accordance with the invention, therefore, the predetermined maximum value Tm of the run-in temperature of the high-pressure turbine at the time t, is reduced by AT,,, to a weighted temperature value Tgew somewhat below Tm. 35 As soon as the rotational speed NI exceeds a rotational speed starting value NST corresponding to the is maximum temperature reduction AT,,,, which occurs at time t2, the difference between T, and the temperature TF at the filter input is proportionally (linearly) reduced. This continues until a rotational speed threshold value NscH is reached at time t3. After this the temperature reduction ATF at the filter input is set to zero. During the first phase of acceleration from t, to a time t4 at which TI reaches the threshold T,, the temperature reduction AT follows the filter input temperature TF with the larger cooling time constant T,. As soon as the temperature difference T,, falls below the temperature difference threshold value TK at the time t4, there is a switch-over to the lower heating time constant T, and the temperature reduction AT is reduced in accordance with the heating time constant T. until it reaches zero.

In this representation the effect of the delay filter of the invention, which works with different time constants for heating and cooling the engine housing, becomes clear in the actual curves for the temperature TI and the rotational speed NI which now exactly reach the predetermined maximum values without overshoot.

The built-in safety logic according to Fig. 1 by way of multiplication of the temperature reduction produced in the rotational-speed-dependent stage 6, by the evaluation factor of the function stage 7, dependent on the temperature difference, has no effect on the embodiment shown in Fig. 3.

Clearly the invention can be applied to any engine in which temperature and running speed need to be precisely controlled, but it is particularly suitable for turbine engines for aircraft.

-is-

Claims

1. A process for controlling the supply of fuel for engines during acceleration and deceleration procedures specifying predetermined maximum values for turbine rotational speeds and temperatures according to engine conditions, in which the differences between these values and the actual values are fed to a regulator (10) for controlling a fuel supply valve the permitted maximum value of the temperature (Tm) is weighted before input into the regulator (10) by applying a temperature reduction (AT) to the maximum value of the temperature (T,,) to produce a weighted temperature value (Tgew Tm - AT) in dependence on the temperature and rotational speed differences as mentioned above and on the heating and cooling behaviour of the engine housing (12) during acceleration or deceleration of the engine (17).

2. A process according to claim 1, in which the run-in temperature of a high-pressure turbine is measured as the actual value of the temperature (T1), and the rotational speed of a high-pressure turbine or compressor is measured as the actual value of the rotational speed (N,,).

3. A process according to claim 1 or 2, in which the temperature reduction (AT) is delayed by means of a filter (4), the filter (4) simulating the heating behaviour of the engine housing (12) with different time constants (.r,, 7-.) for heating or cooling according to different operational states in different flight attitudes.

4. A process according to any preceding claim, in which data corresponding to the said heating and cooling behaviour of the housing are stored in a memory.

5. A process according to claim 4, in which to store the heating behaviour of the engine housing (12) on acceleration of the engine (17) or to store the cooling behaviour of the engine housing (12) on deceleration of the engine (17) the thermal time constants (T,, T.) are determined for different engine states in test runs with supply of heat to, or emission of heat from, the engine housing (12) on transition from one engine state to the other and are then stored in the memory.

6. A process according to any preceding claim, in which the temperature reduction (AT) has an upper limit (AT,,..) and increases proportionally to the difference (ND) between the actual value (NI) of the rotational speed and a rotational speed threshold value (NscH) which is 5 to 20% smaller than the maximum rotational speed (Nm), the temperature reduction (AT) remaining zero for negative differences.

7. A process according to any preceding claim, in which the temperature reduction (AT) increases between a smallest value (ATmin) and a largest value (ATm,x) in proportion to the difference between the maximum value of the temperature (T,,) and the measured actual value of the temperature (TI), and normalised values between 0 and 1 are assigned to this increase, the largest temperature reduction (AT,,,) being equal to the upper limit of the temperature reduction due to the rotational speed difference.

8. A process according to claim 7, in which the increase in the temperature reduction (AT) due to the rotational speed difference (ND) is multiplied by the normalised value resulting from the temperature difference at a multiplication point (M).

9. A process according to claim 8 when appendant to claim 3, in which the output of the multiplication is fed to the filter (4) whose output signal (AT) is subtracted at a first summation point (S,) from the maximum value of the temperature (Tm) to form the weighted temperature (Tgew) -

10. A process according to any preceding claim, in which the regulator is a PID regulator.

11. An apparatus for controlling the supply of fuel for engines, having a temperature gauge for determining the actual value (TI) of the engine and a rotational speed gauge for determining the actual value of the rotational speed (NI) of the engine, a first summation point (S,) for reducing a predetermined maximum temperature (Tm), dependent on flight attitude, by a temperature reduction (AT) and feeding the resultant weighted temperature value (Tgew) to a regulator (10) for the supply of fuel (3) to the engine (17), and a means (4) connected upstream of the first is summation point (S,) for delaying the application of the temperature reduction to the regulator (10) so as to simulate the heating and cooling behaviour of the engine housing.

12. An apparatus according to claim 11, in which the delay means (4) is a first-order filter and applies either a heating or a cooling constant (7,,, T,) in accordance with the position of an adjustment member (5) connected upstream of the filter (4) in dependence on the difference between the maximum value of the temperature (Tm) and the actual value of the temperature (T1).

13. An apparatus according to claim 11 or 12, in which the regulator is a PID regulator, a proportional element (6) is connected downstream of a second summation point (S2) which forms the difference between a threshold value of the rotational speed (Nsoi) and the measured actual value of the rotational speed (NI), the output of this proportional element being conveyed to a multiplication point (M), the difference between the maximum value of the temperature (Tm) and the actual value of the temperature (TI) being formed at a third summation point (S3) connected to a proportional element (7) with a minimum (ATin) and a maximum temperature difference (AT,,..), whose output is coupled to the multiplication point (M), the output of the multiplication point (M) being input to the first-order f ilter (4).

14. An apparatus according to claims 11 to 13, in which the engine is a turbine engine and the temperature gauge (16) and the rotational speed gauge (11) receive the actual values from the high-pressure turbine (13) or the high-pressure compressor (20).

15. A fuel control system substantially as described herein with reference to the accompanying drawings.