The present solution relates to a control system and method for a turbopropeller (or turboprop) engine of an aircraft, based on inlet temperature sensing.
The aircraft 1 includes an airframe 4 defining a cockpit 5; a single operator-manipulated input device (a power, or throttle, lever), 6, and an instrument control panel 7 having a display 8 are provided within the cockpit 5.
The propeller assembly 3 comprises a hub 9 and a plurality of blades 10, extending outwardly from the hub 9. A gas turbine of the turbopropeller engine 2 (here not shown) generates and transmits power to drive rotation of propeller assembly 3, thus generating thrust for the aircraft 1.
As schematically shown in
an axial/centrifugal compressor 12, coupled to an air intake 13;
a high-pressure turbine 14, so called “gas generator”, coupled to the axial/centrifugal compressor 12 via a gas generator shaft 15; and
a low-pressure turbine 16, so called “power turbine”, mechanically decoupled from the gas generator shaft 15 and driven by hot gas expansion.
The propeller assembly 3 is coupled to the gas turbine engine 11 via a propeller shaft 17 and a gearbox 18.
More specifically, the gearbox 18 can include a first gear 18a and a second gear 18b in mesh with the first gear 18a. The first gear 18a can be connected to the propeller shaft 17, in turn coupled to the hub 9 of the propeller assembly 3, and the second gear 18b can be connected to a power turbine shaft 17′, in turn coupled to the low-pressure turbine 16. During operation, the gearbox 18 can step-down a rotational speed of the power turbine shaft 17′, so that a rotational speed of the propeller shaft 17 can be less than the rotational speed of the power turbine shaft 17′.
An actuation assembly 19 is coupled to the propeller assembly 3, to determine the value of a variable pitch angle of the propeller blades 11.
The turbopropeller engine 2 is managed by an electronic control unit 20 (shown schematically in
The turbopropeller engine 2 further comprises: a temperature sensor 22, which is generally arranged within the air intake 13, in order to sense the temperature of engine intake air (that, during operation, flows over the same temperature sensor 22), and is configured to provide a measure of a sensed temperature T1sens.
The sensed temperature T1sens measured by the temperature sensor 22 is relevant to control of the engine operation by the electronic control unit 20; in particular, together with a compressor speed Ng, the sensed temperature 11,ns establishes the position of a Variable Stator Vane (VSV) device coupled to the compressor 12. In a known manner, this device has the purpose of “partializing” the air flow to the compressor 12, so as to avoid a stall condition.
A hydraulic actuator 26, provided with a torque motor, moves a piston 27 back and forth; a mechanical linkage 28 transforms the rectilinear motion of the piston 27 into a circular motion, thereby moving a series of blades 29 of a beta angle β into stator vanes 30 of the compressor 12. The rotation of blades 29 deviates the flow of air in the compressor 12, thus allowing to avoid an excess of air at low speed that would lead to stall of the compressor 12.
The hydraulic actuator 26 is electrically controlled by the electronic control unit 20, in particular by the FADEC, which calculates a beta angular opening (i.e. the value of the beta angle β) according to a corrected compressor speed Ngr. This corrected compressor speed Ngr is the compressor speed Ng multiplied by a parameter that depends on an inlet temperature T2 (that is proportional to the sensed temperature T1sens measured by temperature sensor 22) based on the following expressions:
wherein 288.5K (i.e. 15° C.) is the temperature at ISA (International Standard Atmosphere) condition at sea level.
Since the ambient temperature is inversely proportional to the altitude, during an aircraft descent the ambient temperature will progressively increase (as the altitude decreases), and the engine inlet temperature measurement (the sensed temperature Tlsens) will correspondingly increase. If the sensed temperature Tlsens increases (assuming a constant speed Ng), the corrected speed Ngr decreases.
As shown in
Based on the above, the need is therefore felt for a correct and reliable measure for the sensed temperature T1sens, in particular during an aircraft fast descent transient, allowing accurate control of the engine operations.
The aim of the present solution is to provide an improved control solution for a turbopropeller engine, allowing to meet the above need.
According to the present solution, a control system and a control method are therefore provided, as defined in the appended claims.
For a better understanding of the present invention, preferred embodiments thereof are now described, purely as non-limiting examples, with reference to the attached drawings, wherein:
The present solution stems from the realization, by the Applicant that the temperature sensor 22 may introduce a substantial delay in the temperature sensing, due to the sensor's high time constant τ. For example, the high value of the time constant τ may be due to the sensor construction and/or to the sensor arrangement in the air intake 13 of the turbopropeller engine 2, which may be traversed by a limited flow of air during operation.
The delay introduced by the sensor time constant τ may be modelled by a first order system, as schematically shown in
As shown in
As shown in
As will be discussed in the following, an aspect of the present solution thus envisages a suitable compensation of the sensed temperature T1sens provided by the temperature sensor 22, in order to compensate for the delay introduced by the same temperature sensor 22.
As shown in
By a proper choice of the lead value τa, it may be possible to achieve a direct compensation of the lag on the sensed temperature T1sens, generating a compensated temperature T1comp.
The present Applicant, however, has realized that this solution suffers from a drawback due to the high value of the time constant τ, causing an amplification of the little variations introduced by the noise on the sensed temperature T1sens, reducing the useful signal.
In particular, if the difference between τ and τa is very high (since the value of the time constant τ is high), the lead compensator block 32 behavior is the same as a derivative block, thus amplifying the noise; this same noise deteriorates the quality of the signal, producing small variations on the VSV command.
Accordingly, an aspect of the present solution envisages the estimation of the delay and the error affecting the measure of the sensed temperature T1sens by the temperature sensor 22, in particular during the aircraft descent, using the measure of a different sensor (as will be discussed in the following, a sensor providing altitude information, in particular a pressure sensor) to produce a suitable compensation quantity.
In particular, assuming that the time constant τ of the temperature sensor 22 is known, it is possible to estimate the delay and the error affecting the temperature measure during the aircraft descent.
In detail, the dynamic error e(s) between the sensed temperature T1sens and the actual temperature Tlact can be expressed as:
e(s)=T1sens(s)−T1act(s)
Considering, as discussed above, the temperature sensor 22 as a first order lag model, the sensed temperature T1sens can be expressed as:
so that the above expression for the error e(s) may be expressed as follows:
It is possible to estimate the error e(s) in steady state during a constant fast descent, i.e. considering the rate of descent as a constant value A (ramp descent).
Under this assumption, the actual temperature T1act can be expressed, in the time domain, as:
T1act(t)=At·u(t)
and, in the Laplacian domain, as:
Considering the final value theorem:
Accordingly, the steady state error during a descent depends on the rate of descent A and the sensor lag (time constant τ).
Moreover, the rate of descent A can be expressed as:
so that the above expression for the steady state error can also be formulated as:
In order to determine the rate of descent A, the hypothesis can be made that the actual temperature T1act is given by the sum of the temperature calculated by ISA (International Standard Atmosphere) condition, i.e. the ISA temperature T1ISA, and an additional value ΔTday, associated to the day condition and that does not depend on the time of descent (in other words, it can reasonably be assumed that this additional value ΔTday is constant with time and as the altitude decreases):
T1act(t)=T1ISA(t)+ΔTday.
The ISA temperature T1ISA indeed increases as the altitude decreases, as shown by the detailed values reported in a Table according to the International Standard, shown in
From the above expression (and considering that the additional value ΔTday is constant with time), it follows that:
It follows that it is possible to calculate the descent rate A based on the knowledge of the ISA temperature T1ISA.
As shown in the same Table of
In this respect,
The aircraft 1 is provided with a pressure sensor, denoted with 35 in the following (see
The error e(s), given, as discussed above, by the following expression:
can thus be compensated by adding, to the sensed temperature Tsens, the following compensation quantity comp(s), equal and opposite to the above error expression:
wherein the ISA temperature TISA is determined as a function of the external pressure P0 measured by the pressure sensor 35:
T1ISA=f[P0]
(it is noted that the equivalence sT1act=sT1ISA has been exploited in the above expression).
In other words, adding the compensation quantity comp(s) to the sensed temperature Tsens at the output of the temperature sensor 22 allows to fully compensate the error e(s) (i.e. e(s)=0).
The compensation system 40 includes:
a first calculation block 42, which receives at the input altitude information (in the example shown in
a second calculation block 44, coupled to the first calculation block 42, which receives at the input the calculated ISA temperature T1ISA and generates the compensation quantity comp, implementing the transfer function sτ/(1+sτ) (corresponding to a derivative filter);
an adder block 45, having a first adding input coupled to the second calculation block 44 and receiving the compensation quantity comp, a second adding input receiving the sensed temperature T1sens (generated by the first order lag block 30 modelling the temperature sensor 22, with transfer function 1/(1+sτ), starting from the actual temperature T1act), and a sum output providing the compensated temperature T1comp (wherein the error e(s) has been compensated).
This compensated temperature T1comp, as shown schematically, can then be used by the electronic control unit 20 to implement control operations on the turbopropeller engine 2, such as for controlling the opening/closing state of the VSV device 25 as previously discussed in detail, thereby providing a complete control system 50.
an anti-noise filter block 46, in particular a second order low pass filter, inserted between the first calculation block 42 and the second calculation block 44, configured to filter and limit the bandwidth of the signal indicative of the external pressure P0 provided by pressure sensor 35, so as to improve the signal-to-noise ratio and aliasing effect (also considering that, since the rate of descent is not very fast, the bandwidth required for the compensation signal is generally small); and
a saturation block 47, at the output of the second calculation block 44, configured to allow the compensation only during a descent of the aircraft 1.
During an ascent, the output of the saturation block 47 (representing the compensation quantity comp for the adder block 45) is always set to a 0 (zero) value, so no compensation is provided; during the descent, the output of the saturation block 47 is limited (saturated) to a saturation value T1sat, for example equal to 30K (this value can be tuned depending on the time constant τ), in order to determine a maximum value for the compensation quantity comp and avoid an over compensation.
a derivative block 44′, implementing a derivative function of the ISA temperature TISA, according to the expression τ(dT1ISA/dt); and
a first order lag block 44″, implementing a lag function according to the expression 1/(1+sτ).
The advantages of the present solution are clear from the previous discussion.
In any case, it is again underlined that the present solution provides an effective system to compensate for the delay introduced by the temperature sensor 22, allowing to achieve improved engine control operations and reduced operating and maintenance costs.
Advantageously, the disclosed solution exploits altitude information provided by a different sensor to calculate a compensation quantity comp, that is used to compensate for the error on the sensed temperature T1sens measured by the temperature sensor 22 (thus implementing a “sensor fusion” algorithm). In particular, an external pressure sensor 35 is used to calculate the compensation quantity comp, the pressure sensor having a much quicker response (and a much lower time constant) than the temperature sensor 22.
Finally, it is clear that modifications and variations can be made to what is described and illustrated herein, without thereby departing from the scope of the present invention as defined in the appended claims.
In particular, it is underlined that, although generally applied to a fixed-wing aircraft, the present disclosure may further apply to rotary-wing aircraft, tilt-rotor aircraft, or other apparatuses including a pitch-changing propeller assembly and a gas generator coupled to the aircraft.
Number | Date | Country | Kind |
---|---|---|---|
18425044.7 | Jun 2018 | EP | regional |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2019/066098 | 6/18/2019 | WO | 00 |