GB2218537A - Engine control - Google Patents

Description

ENGINE CONTROL 2218537 The invention relates to an engine control unit for

a L-urbomachine which is operative to control the level of -Fr-.e'j- flow to achic-ve a-.-, engine speedl reqiii red by an input demand.

In particular, the invention concerns the arrangement and operation of a controller for an aircraft gas turbine, the controller itself comprising, basicall-,;,, an electronic control unit or a suitably programmed computer. Controllers of this type exercise dynamic control over the engine by monitoring a number of critical variable parameters such as engine speed, turbine temperature, compressor delivery pressure etc. and, in accordance with a pilot's' speed input demand, calculating the rate of fuel flow required by the engine. When the pilot demands increased engine speed the controller calculates a level of overfuelling required to accelerate the engine. In this context and in the remainder of this specification engine deceleration is to be regarded simply as engine acceleration in the negative sense. An overfuelling requirement therefore, is be similarly construed to include a fuel reduction as appropriate.

When an engine is accelerated slowly the relationship of compressor delivery pressure and turbine temperature to engine speed remain fairly closely in line with the corresponding values at constant engine speed.

However, if acceleration is allowed to reach too high a value compressor stall may occur and lead to an engine surge. Normally sufficient margin exists between the engine running line and the surge boundary, but whIen the excess fuel necessary to produce engine acceleration is burned the engine pressure rat:-o is driven towards the surge region. If the level of excess fuel, or overfuelling, is too high relative to the dynamic characteristics of the engine a surge can result.

The engines of civil aircraft are rarely, if ever, operated so close to the limits of their performance as to render them susceptible to surge, but this is not so in military aircraft. In order to fully exploit the dynamic characteristics of an engine. it is frequently driven to the limits of its performance envelope and surge problems are a constant problem. Incipient surge sensing would be the ideal but a single engine parameter exhibiting a unique change as stall or surge conditions are approached has not been discovered and it has proved almost impossible to predict the onset of surge by noting the changes in balance between two ormore parameters. The only practicable solution, therefore, is to eirploy a form of open-loop control which strikes a balance between engine performance over a wide range but maintains an adequate safety margin over surge conditions.

Recognising that the surge boundary does not follow a linear relationship, when pressure ratio is plotted against engine speed, the present invention has for an object to provide an overfuelling characteristic having a non-linear slope and which is dynamically compensated to reflect the slope of the surge boundary.

According to the present invention an engine control unit for a turbomachine operative to control the level of fuel flow to achieve an engine speed required by an input demand includes means for solving an engine speed control function in which a maximum value is imposed upon a permitted engine acceleration rate, said maximum value is substantially reduced subject to certain monitored variables having values outside a predetermined range of values.

In one embodiment of the invention when engine speed is less than a threshold value and when an acceleration is demanded the normal maximum acceleration limit is reduced by 50% initially to obviate a surge condition and is subsequently returned progressively to the maximum limit as engine acceleration builds-up and the engine is capable of coping with the overfuelling requirement.

The invention and how it may be carried out in practice will now be described in greater detail with reference, by way of example only, to the accompanying drawings in which:

Fig 1 is a schematic illustration of an engine control system Fig 2a shows a curve of fuel flow against engire speed including upper and lower limits Fig 2b shows the relationship between fuel flow and control current Fig 3 shows a logic dlagram for an acceleration control algorithm Fig 4a shows a logic diagram for an alternative algorithm, and Fig 4b shows the effective logic of Fig 4a The principle of the invention is to reduce an acceleration demand initially until the rate of engine acceleration has reached a level at which the engine is able to cope with the amount of overfuelling required to produce the desired accele=tion rate.

The engine control system of Fig 1 embodies the principle of NH dot control, where NH is defined as the engine speed, or in a multi-spool engine as the speed of the high pressure spool, and NH dot is the rate of change of NH or engine acceleration. The engine itself is indicated diagrammatically by the block 2, the main engine control unit is represented at 4 and the fuel system at 6. The amount of fuel 8 supplied to the engine 2 is controlled by the fuel SYS-Lem 6 in accordance with a level of control current 10 provided by the output of control unit 4.

Control unit 4 operates closed loop control of engine speed NH subject to performance limitations based on predetermined limits for turbine blade temperature TBT and engine acceleration NTH dot. Instantaneous values of these parameters are fed back from engine C> - Lransducers to summing points 12 14 and 16 respectively for comparison with TBT datum and NH dot datum values and an NH demand signal generated by the setting of the pilots speed demand lever.

Basically, at summing point 16 the engine speed feedback signal is compared with the pilot's NH demand signal to provide an NH error which the control unit processes through lowest wins type logic together with the performance limit datums to generate a control current 10 which sets the level of fuel flow via fuel system 6.

i 1 j1 At a steady running speed, see Figs 'La and 2b,NH error J s substantially zero and control unit 4 generates P control current 10 the value of which corresponds to the steady state engine fuel flow requirement. This value varies with engine speed and around thp aircraft flight envelope, with power off-take and it varies between engines and fuel systems which have to be matched to a particular engine. A typical relationship between fuel flow and the control current is shown in Fig 2b. The fuel flow has maximum and minimum limit values denoted at AM and DCU respectively, the exact values of which are a function of engine speed. At any given engine speed NH, however, there are maximum and minimum fuelling limits which in this example occur at 100 mA and 900 mA respectively; between those limits fuel flow is inversely proportional to control current As shown in Fig 2a to maintain conditions at progressively correspondingly greater levels of steady running higher speeds fuel f low are required, although the relationship is not linear.

Similarly, at higher engine speeds greater amounts of fuel are required to produce a given rate of engine acceleration.

progressively, Thus, to accelerate an engine from given speed a level of overfuelling is supplied commensurate with the maximum permitted limit. The level of fuel supplied is reduced as far as the DW limit if required t6 decelerate the engine. The limits shown in Fig 2a are based upon the normal design flight envelope within which consistent and reliable engine performance can be expected.

At e-xtreme limits unstable 9 however, engine performance can become or at least unpredictable. As mentioned above turbomachines may be susceptible to -A- z surge problems during early stages 0 rapid accelerations from low speed, reslam accelerations and when reducing speed right back to flight idle, for example.

The invention proposes to obviate these problems by altering-initially the maximum rate at which the engine can accelerate. That is, the maximum level of overfuelling is limited to a level with which the engine is able to cope at low speeds. Obviously, these levels are not the same in different engines so that precise figures mentioned below and included in the accompanying drawings are to be regarded in a very general way and as very approximate only.

According to the invention the control unit 4 will automatically implement at relatively low engine speeds an iterative algorithm to limit the acceleration rate initially, when acceleration rates higher than a predetermined maximum are demanded. Fig 3 shows in an equivalent logic gate type format the logical processes and decisions involved in implementing a two gradient acceleration control algorithm.

The control algorithm of Fig 3 is designed to modify the acceleration control loop in the reslam acceleration case and to avoid acceleration delavs except in circumstances in which the engine is susceptible to surge.

The NH dot datum is reduced therefore only during a demanded deceleration or when the control current 10 is at or close to the DCU limit, ie. is greater than 800 mA, provided two further conditions are also satisfied.

-V i i Firstly, that the engine speed is between defined limits 60Z and 85% of maximum speed, and airspeed is below a predetermined le-vel defined by the freestream pressure Pto less or equal to 60 Kpa (Kilopascals).

Once a "true" logic output has been set the output is latched, by the feedback loop labelled "latching signal", to maintain the reduced datum until an acceleration is detected, at a rate greater than 0.5% NTH per sec, or the airspeed or the engine speed exceeds the predetermined limits mentioned above.

When the output is unlatched and returned to its normal state the NH dot datum value is returned to its original schedule value. Preferably, this is not done in step fashion but it is ramped back to its full value over a short period of time, say, of the order of one second in the example described.

The engine control unit embodying the invention being. described comprises a microprocessor based digital control system. The control algorithms are all embedded in software programs and sub-routines. The main engine control program is executed cyclically so that all relevant inputs are periodically sampled and the control functions re-evaluated. The control outputs are therefore also refreshed periodically. The period of a typical cycle of operation is of the order of mS The acceleration control algorithm described above is contained in one such sub-routine, or suite of sub-routines, and is called-up when the initial input conditions for acceleration limitation are fulfilled.

It follows therefore, that the algorithr. will be executed in a number of successive cycles over a period of time, also, that the NH dot datum return ramp function will be spread over a number of successil.re cycles.

7 the ramp funct During performance o'L 1-ion the NH dot datum value may be stepped back to the full scheduled value in a series of increments. Relatively coarse incremented level may be maintained for several control program cycles, alternatively, finer step increments may be incremented in each cycle so that the full schedule value is arrived at in a series of iterations.

The algorithm represented in Fig 3 can also be adapted to implement a sequence of stepped reductions of the NH dot datum value. Thus, the acceleration rate may be limited to one or more further reduced values in intermediate steps before being returned to the scheduled value.

In a further embodiment of the invention a somewhat simpler algorithm represented by the equivalent logic gate diagrams of Fig 4a. At any time, if the engine is not accelerating at a rate greater than 33% of the scheduled NH dot value then the corresponding datum value is reduced subject to engine speed NH being measured within defined limits, a lower limit of 60% of maximum NH and a -variable upper limit which is a function of Pto, which is, in turn, related to airspeed.

The NH dot datum is also reduced when the fuel supply control current exceeds a predetermined threshold, 750 mA in the example, and engine speed is between the defined limits.

1 ZY However, if the control unit output is limited by the DCU limit then, in the example, the actual engine acceleration rate NH dot which can be achieved is bound to be less than 33% of the scheduled datum value, 1ding the control unit Still I-as control of the prov L - I L engine. The effective logic is therefore simpler and reduces to the form illustrated in fig 4b.

In this modified arrangement latching of the output is unnecessary and the algorithm will operate in respect of ann.;, acceleration arising within the defined speed limits. Upon reaching 33% of the schedule NH dot datum.,,Talue or exceeding the defined speed limits the reduced NH dot datum is returned progressively to the scheduled value over a period of time the length of which is a function of Pto. In a system in which the control function is evaluated cyclically this progressive return to schedule value is effectively an iterative process.

Again, although not illustrated in Fig 4a, intermediate levels of reduction of the NH dot datum -may be implemented according to various progressive thresholds being achieved in any of or combination of the controlling input parameters.

Claims (10)

1. An engine control unit for a turbomachine operative to control the level of fuel flow to achieve an engine speed required by ail input demand includes means for solving an engine speed control function in which a maximum value is imposed upon a permitted engine acceleration rate. said maximum value is substantially reduced subject to certain monitored variables having values outside a predetermined range of values.

2. An engine control unit as claimed in claim 1 wherein one of the monitored variables is the fuel supply control signal and the maximum permitted engine acceleration rate is reduced when said fuel supply control signal is close to a fuel limit.

3. An engine control unit as claimed in claim 1 or claim 2 wherein one of the monitored variables is air pressure in an air intake to the turbomachine and the maximum permitted engine acceleration rate is reduced when said air pressure is less than a predetermined value.

4. An engine control unit as claimed in any preceding claim wherein one of the monitored variables is engine speed and the maximum permitted engine acceleration rate is reduced when the engine speed is between predetermined limits.

41 1 2218 537

5. An engine control unit as claimed in any preceding cl laim. comprising a microprocessor operatively controlled by software including a program arranged to execute an a l go r ithirt to limit permitted engine accp-leration_ in accordance with the _reduced maxil-..1'-m value when the values of the monitored variable lie outside said permitted ranges of values.

6. An engine control unit as claimed in any preceding claim wherein acceleration is limited by reducing the reference datum input of an error in an engine acceleration control loop.

7. An engine control unit as claimed in claim 6 wherein the reduced reference datum input is returned to full value in accordance with an iterative algorithm.

8. An engine control unit as claimed in claim 77 wherein the reduced reference datum input is returned to full value progressively, over a period of time.

1

9. An engine control unit as claimed in claim 7 or 8 wherein the reduced reference datum input is returned to full value progressively in a series of incremental steps.

10. An engine control unit substantially as described with reference to the accompanying drawings.

Published 1989 atThe Patent Office, StELteHouBe,66!71 High Holborn, London WCIR4TP. Further copies maybe obtainedfrom ThePatentoffice. Sales Branch, St Mary Cray, Orpington, Kent BR5 3W). Printed by Multiplex techniques ltd, St Mary Cray, Kent, Con. 1/87