CHAPTER 12

CONTROL AND FLY-BY-WIRE

Control and FBW

TABLE OF CONTENTS

12.1. INTRODUCTION.........................................................................3
12.2. CONCEPTS AND DEFINITIONS.......................................................4
12.2.1. Systems................................................................................. 4
12.2.2. Models..................................................................................5
12.2.3. Control.................................................................................. 5
12.3. MATHEMATICAL TOOLS...............................................................6
12.3.1. Differential Equations................................................................6
12.3.2. Laplace Transforms...................................................................7
12.3.3. Complex Numbers..................................................................12
12.4. ANALYSIS OF SYSTEMS AND CONTROL.........................................12
12.4.1. Transfer Functions..................................................................13
12.4.2. Block Diagram Algebra.............................................................14
12.4.3. Stability and Performance.........................................................16
12.5. CONTROL SYSTEMS APPLICATIONS.............................................29
12.5.1. Flight Control Systems.............................................................30
12.5.2. Fly-by-wire concepts...............................................................31
12.5.3. Other applications...................................................................32
12.6. TESTS AND CERTIFICATION........................................................34
12.6.1. Ground and Flight tests............................................................34
12.6.2. Certification aspects................................................................35
12.7. REFERENCES.........................................................................35
APPENDIX 12A - EXERCISES

Control and FBW

12.1.

INTRODUCTION

The flight testing activity demands a constant contact with complex systems. In
fact, a complete aircraft can be considered a system of systems, according the
concepts presented in this text.
An auto-throttle, a hydraulic system (operating landing gear or brakes), the
flight control system (ex: using fly-by-wire concepts), auto-pilots, or a FADEC to
engine control, to all of them, the system theory and control must be applied.
A control system consisting of interconnected parts is designed to achieve a
desired purpose/objective from the controlled system. To understand the
purpose of a control system, it is useful to examine examples of control
systems through the course of history. These early systems incorporated many
of the same ideas of feedback that are in use today.

18th Century James Watts centrifugal governor for the speed control of a steam engine.
1920s Minorsky worked on automatic controllers for steering ships.
1930s Nyquist developed a method for analyzing the stability of controlled systems
1940s Frequency response methods made it possible to design linear closed-loop control
systems
1950s Root-locus method due to Evans was fully developed
1960s State space methods, optimal control, adaptive control and
1980s Learning controls are begun to investigated and developed.
Present and on-going research fields. Recent application of modern control theory includes such
non-engineering systems such as biological, biomedical, economic and socio-economic systems.
For this wide range of applications, there have been developed new techniques and theories such
as adaptative, robust, predictive controls, etc.
Modern control engineering practice includes the use of control design
strategies for improving manufacturing processes, the efficiency of energy use,
advanced space vehicles control, including weapons, among others.
We also emphasize the notion of a design gap. The gap exists between the
complex physical system under investigation and the model used in the control
system synthesis.

Control

The iterative nature of design allows us to handle the design gap effectively
while accomplishing necessary tradeoffs in complexity, performance, and cost
in order to meet the design specifications.
This course will be restricted to furnish the general idea how to analyze the
systems and how to apply linear control to them. More advanced applications
requires courses that are more sophisticated.
The Linear Control Theory deals in modelling physical systems, describing them
with block diagrams and dynamic equations that define their behavior. The
control system is designed to obtain responses of a general system with
determined characteristics of performance and stability.
For further and more advanced information, a specialized bibliography is listed
at the end of the chapter. Also a list of exercises should orient the student to
the exam.

12.2.

CONCEPTS AND DEFINITIONS

Before we begin the study of analysis and synthesis of control systems, there is
the need for an objective concept and clearly define the terms relating to the
systems and other related terms.
Despite restricted the theory herein, the presented principles and
fundamentals remain the same and are the basis for the study of complex
system.

12.2.1. Systems
SYSTEM = A set of elements/devices, selected and organized, interconnected
for a desired purpose. Every system is delineated by its spatial and temporal
boundaries, surrounded and influenced by its environment, described by its
structure and purpose and expressed in its functioning.
PROCESS = The device, plant, or system under control. The input and output
relationship represents the cause-and-effect relationship of the process.
Another important characteristic is the architecture that defines the
relationship among the internal parameters.
PARAMETERS = Important factors used in the model to define the process
under analysis. Defines the dynamic between input/output.
VARIABLES (INPUT/OUTPUT) = Important factors defined to excite the process
or the system (input); outcome of the system in response to the inputs
(output). Depending on the modeling, outside variable can be considered
(disturbances).
4

Control and FBW

12.2.2. Models

Linear or not-linear superposition principle

Variants or not-variants time dependent
Analogic or digital infinite possible values for variables
Continuous or discrete time interval (ex: sampling)

Mathematical, transfer function or blocks diagram Different types of

representation of system dynamics and relationship among variables.
Dependent of architecture and parameters.

12.2.3. Control
Classic or Modern
Classical control theory deals with solving differential equations in a frequency
domain (Laplace/Fourier/Z-transform), while in modern control theory, we solve
in the state space (time-domain). In classical control we can analyze SISO
(single input, single output) systems while a state space based approach is
better suited for MIMO systems and can handle non-linear & time invariant
systems as well (which is how most practical problems are).
That being said, the state space approach (hence modern control theory) is not
something relatively new, but gained popularity after the 60's because it
involved numerically (and computer) friendly matrix operations.
The need to move further from simple controllers was first posed by aerospace
researchers due to the MIMO systems that they wanted to control, this paved
way for modern control theory.
Open loop

Closed loop

Control

12.3.

MATHEMATICAL TOOLS

Mathematics is playing an increasingly important role in all sciences, not only

in engineering, inciting clouding boundaries between classical techniques and
computational capabilities. Remarkable progress has been made in both theory
and applications of all-important areas of control theory, which reaches far
beyond the engineering.
Complex systems require different approaches for different problems.
Therefore, we present here some classical tools that are the basics for Control
Theory, classic or modern.

12.3.1. Differential Equations

Differential equations are nothing more than a man's attempt to model
phenomena of the nature
Differential equations are equations where the unknowns parameters are
functions. They are presented in the form of algebraic equalities involving
differential or derivatives.
Example:

This differential equation x(t) is the unknown function. It represents the

movement of the mass shown below when excited by an oscillating force given
by F = F0 cos(t).

Figure 12-1 MASS SPRING DAMPER SYSTEM

Control and FBW

The equation of motion governing the mass-spring-damper system is an example of

linear ordinary differential equation with constant coefficients. Ordinary differential
equations have the incognita function with one free variable, in this case is variable
time (t). Otherwise, it is said partial differential equation. Structural systems subjected
to dynamic vibration modes are partial differential equation (time and directions).
Specific methods of solution of differential equations are not on the scope of this
course.

12.3.2. Laplace Transforms

Let f(t) be a real function with the real variable t > 0. The Laplace transform of
f(t) is defined by:

were s is a complex variable, named complex frequency:

s = + i, with real numbers and .

One important application to this tool is to solve differential equations. In

practice, it is not necessary always to calculate the Laplace transforms through
the definition. The transforms are tabulated and with some properties we can
solve most of the problem analysis and synthesis of control systems.
The essential properties and theorems are the following:
Multiplication by real constant;

Addition and subtraction;

Differentiation;

Integration;

Control

Time displacement;

Frequency displacement;

Initial value theorem;

lim f ( t ) =lim sF (s)
t0

Final value theorem;

lim f ( t ) =lim sF (s)
t

s 0

Complex translation;
e

+/-at

f(t) = F (s -/+ a)

Complex convolution; and

Time convolution.
Let F(s) and G(s) denote the Laplace transforms of f(t) and g(t), respectively.
Then the product H(s) = F(s).G(s) is the Laplace transform of the convolution of
f(t) and g(t) and is denoted by h(t) = (f*g) (t), and has the integral
representation

The opposite way to solve the complete problem can be accomplished using
the inverse Laplace Transform, defined by:

Control and FBW

Likewise, it is not necessary to use the definition to solve problems of this

nature. Just use the table and the partial fractions expansion method.

Figure 12-2 LAPLACE TRANSFORMS TABLE

Control

EXAMPLE 1: Show the Laplace transform as seen on the table.

Equating real and imaginary parts,

EXAMPLE 2: Find the inverse transform of:

F ( s )=

2
2
s +4 s + 3 s
3

The denominator is (s3 + 4s2 + 3s) = s.(s+3).(s+1)

Then F(s) = A/s + B/(s+3) + C/(s+1)
With: A = 2/3
B = 1/3
C = -1
Finally, f(t) = 2/3.u(t) + 1/3.e-3t e-t

The final observation about the solution of a differential equation is made by

looking the answer f(t). The solution of the homogeneous equation is called
free response (dynamic of the system without external excitation). On the
other hand, there is a particular solution, corresponding to the forced
response of the system. Thus, the total solution is given by:

y(t) = free response + forced response.

10

Control and FBW

Another way to interpret the total dynamic response of a system is to divide as

follow:

y(t) = transient response + steady state response

12.3.3. Complex Numbers

A complex number is a number with the following form:
a + bj , where j = imaginary number ( 1 ), named CANONICAL FORM, or in
the EULER FORM as M . e j =M ( cos ( ) + jsen ( ) ) also called POLAR FORM.
As can be easily seen, a complex number can be interpreted as a vector and
therefore, all the rules for their operation are the same as vector calculus.

EXERCISE: Prove that the relation below gives the multiplication of complex
numbers:
(a + bj) . (c + dj) = (ac-bd) + (bc+ad)j
To complement, show how this multiplication is given in polar form.

12.4.

ANALYSIS OF SYSTEMS AND CONTROL

The rational interpretation of nature gave rise to models, particularly

mathematical models, to describe the world around us. This description is
given in the form of equations as set out by physical laws. Particularly,
differential equations describe the dynamics of systems.
Thus, most of the systems, whether mechanical, electronic, electromechanical,
thermal, hydraulic, etc. can be characterized by these differential equations.
The response of a dynamic system to an input (or an excitation function) may
be obtained thru the solution of a set of differential equations that characterize
the system. This solution is the output model.

11

Control

Figure 12-3 SYSTEM DESCRIPTION (DIFFERENTIAL EQUATIONS)

The objective of a control system is to get a desired output of the physical
system, thru adequate inputs regardless outside perturbations or gaps in the
modeling.
In the next items, will be presented the tools and the methodology to reach this
goal.

12.4.1. Transfer Functions

The concept of Transfer Function of a system is the relationship between the
input and the output, as defined below:

G(s) = Transfer Function =

output
input

Y ( s)
X ( s)

(zero initial conditions)

Lets point out some important properties:

The transfer function results from the Laplace transform of the differential
equation representing the system, considering zero initial conditions.
Therefore, Y (s) is the forced response from the system.
If you apply a unitary impulse (t) at the entrance of a system, the Laplace
transform of the output is the very system transfer function (show that).

Considering G(s) in the form:

G(s) =

K ( sz 1 ) ( sz 2 ) ( szm)
, the nomenclature to be used is:
( s p 1 ) ( s p 2 ) (s pn)

z1, z2,.,zm (roots of the numerator) = ZEROS OF THE SYSTEM.

p1, p2,pn (roots of the denominator) = POLES OF THE SYSTEM.
(later will be shown that the poles are related with the stability of the system)

12

Control and FBW

Figure 12-4 TRANSFER FUNCTION BLOCK DIAGRAM

12.4.2. Block Diagram Algebra

A physical system to be controlled and your control system may be constituted
by several components. In order to show the functions performed by these
components, it is usual to represent them thru FUNCTIONAL BLOCKS. These
blocks have equations that represent their dynamics and therefore their
respective transfer function. The relations between blocks are modeled so a
block diagram can graphically represent the system.
In this way the block diagram showing the functions performed by the various
system components and the interrelationship between them. Also show signs
flows (variables) and therefore the changes that this signal undergoes while
going through the various system components.
We have seen that there are two major categories of systems: OPEN AND
CLOSED LOOPS. Similarly, the corresponding representations are shown in the
figure below. The representation showed of closed loop transfer function (CLTF)
is called "canonical form".

Figure 12-5 OPEN LOOP AND CLOSED LOOP BLOCK DIAGRAMS

Where:
X(s) = input signal
Y(s) = output signal
G(s) = transfer function (direct)
H(s) = feedback transfer function
13

Control

E(s) = error signal

B(s) = feedback signal
Normally, block diagrams are not present in simple shapes shown. Complex
systems have different functions and therefore a diagram with various
interrelationships between signals and variables occur. Several "summation
points" and blocks in "cascade" are the rule rather than the exception.
However, it is possible to reduce them to simple shapes, which are equivalent
and condensed through algebraic manipulation of transfer functions. This
method is called block diagram algebra.
To simplify our work, some forms have already been calculated and tabulated.

Figure 12-6 RULES OF BLOCK DIAGRAM ALGEBRA

14

Control and FBW

Figure 12-7 EXAMPLE OF PARTIAL BLOCK DIAGRAM (from Nonlinear Dynamic

Inversion Baseline Control Law: Architecture and Performance Predictions Christopher J. Miller AIAA paper)

12.4.3. Stability and Performance

The general procedure in the analysis of a control system is as follows:

1 - Determine the transfer function of each system component;

2 - Mount block diagram;
3 - Determine the main characteristics of the system.

The main items that feature a control system:

STABILITY
TRANSIENT RESPONSE
STEADY - STATE ERROR
FREQUENCY RESPONSE

In classical control theory, there are several methods for determining the
characteristics of a linear system:
- Direct solution of differential equations of the system;
- Root Locus (LGR);
- BODE plot of frequency response;
- NYQUIST plot of frequency response;
- NICHOLS chart of frequency response.

Let's see some applications of these methodologies.

STABILITY

15

Control

The tendency of a system to develop restoring forces equal to or greater than the
disturbing forces to maintain the state of equilibrium is known as stability.

A system is said to be stable if the output follows the input in a finite time. On
the other hand, it is considered unstable if the output does not follow the
input oscillating continuously or diverging to infinity, out of control. Stability
can be analyzed according to two aspects: absolute or relative stability.

Figure 12-8 SYSTEM BEHAVIOUR WITH RESPECT TO STABILITY

ABSOLUTE STABILITY = simply classify the system with respect to stability,
without quantifying. The absolute stability depends only of poles locations
(roots of the denominator of the transfer functions).
Analyzing the response (inverse Laplace transform), that is a sum of
exponentials functions. Therefore, if all F(s) poles have negative real part, the
transient response will tend to zero and the system will be stable. Routh
stability criterion is the mathematical procedure to verify this pole location
(bibliographic reference [1]).
RELATIVE STABILITY = this concept worries with the level of stability of the
system. Quantification in this case, is determined through concepts such as
GAIN MARGIN and PHASE MARGIN, which will be seen below. These values are
determined with the tools named techniques in the frequency domain.

16

Control and FBW

Figure 12-9 GAIN MARGIN AND PHASE MARGIN BODE PLOT

FIRST AND SECOND ORDER SYSTEMS

First order systems has a simple behavior, shown in figure below. The
performance characteristics will be analyzed later. In time domain is a simple
exponential function.

Figure 12-10 FIRST ORDER RESPONSE TO STEP INPUT

In the study of control systems, differential equations of 2nd order are very
important because, higher order systems can often be approximated by such
systems.

Above, a very usual differential equation form, where y is the output:

a=1
b = 2 n
c = n2
x(t) = (input)
= damping coefficient
n = natural frequency (>0)
In this case, the poles of the transfer function, Y(s) / X(s) are given by:
s = - n

( 21)
17

Control

For second order, the response to step input is dependent of the poles location.
In other words, the parameter .
EXERCISE: Analyze the possibilities
for and stability.
Now let us make some considerations on the complex roots of characteristic
2nd degree equation from the system and the shape of the resulting response.
Assuming the roots in the form a + bj, the response-function takes the form:
y (t) = c1 e(a+bj)t + c2 e(a-bj)t
It is to show that that the above expression may be written:
y (t) = K1 + K2 eat sin(bt)

K1 and K2 are constants

EXERCISE: Analyze the possible cases and responses formats. (remember

symmetry).

Figure 12-11 SECOND ORDER RESPONSE TO STEP INPUT

TRANSIENT RESPONSE
The transient response was defined as part of the response that tends to zero
as time goes to infinity. This response is rich in information about the system
and about it are defined various parameters commonly used for evaluation and
control system specification.
It was standardized a step input as excitation for transient analysis thereat, is
easily generated and produces an output well defined in the control system. As
excitation function, can also be used the unit impulse function and sometimes
the unit ramp function. These evaluation options, however, are less common
and depend on the studied situation.

18

Control and FBW

Are defined factors that measure the system PERFORMANCE in transient phase.
These parameters depend on the system order and are used, in the design, to
provide the so-called "specifications in the time domain".

ZERO ORDER SYSTEMS

Y ( s)
X ( s)

= K (constant)

=>

thus y(t) = K u(t)

Figure 12-12 ZERO ORDER RESPONSE TO STEP INPUT

FIRST ORDER SYSTEMS

Y ( s)
X ( s)

K
s+ a

=>

thus y(t) =

1
Ka

(1 -

eat

19

Control

Figure 12-13 FIRST ORDER RESPONSE TO STEP INPUT

SECOND ORDER SYSTEMS

Y ( s)
X ( s)

K 2n
s 2+2 n s+ 2n

=> thus y (t) = K1 + K2 eat sin(bt), where, K1 and K2

are:
K1 = constant
K2 =

k
1 2

a=-

b = d = (n

1 2

20

Control and FBW

Figure 12-14 SECOND ORDER RESPONSE TO STEP INPUT

A 2nd order control system is specified based on a number of parameters

resulting from the response to unit step of the system, called quality criteria.
These indicators, shown in the figure above, are:
Overshoot biggest value reached at transient time, Mp (%);
TR ou TR1 RISE TIME time needed from 0 to 100% input or from 10% to 90% input;
TP PEAK TIME necessary time to reach the first peak value of response (overshot
time);
TS SETTLING TIME required time for the response be stabilized within a
predetermined band, around the unit.
ess STEADY-STATE ERROR stabilized value versus the input (difference)
The previous figure 12-11 shows the influence of the parameter of the system
at the response.
After some mathematical manipulation, it is possible to achieve the following
relations for the performance in time domain as function of the parameters
and n:
TS

1+1,1 +1,4
n

Mp = 100 eA ,

where A =

1 2

21

Control

TS

TP =

3
n

where d = n

1 2

STEADY-STATE PERFORMANCE

After the analysis of the transient system performance, we access the

characteristics in steady state. Consider the closed loop system shown below.
The error signal E(s) acting on this system has its corresponding time domain
e(t). We can show (exercise) that E (s) can be written as:
E ( s )=

X (s)
1+G ( s ) H ( s)

Figure 12-15 CLTF

The error in the steady state is defined as:

e ss= lim e (t )
t

The error in the steady state is resultant from the interaction of two dynamics:
the input type and the actual system dynamics, i.e., the degree of the
denominator of the product G(s) H(s). It defines the error constants as follows:
22

Control and FBW

Kp (position error constant) =

Kv (velocity error constant) =

lim G ( s ) H (s)

s0

lim s G ( s ) H ( s)

s0

Ka (acceleration error constant) =

lim s2 G ( s ) H (s)

s0

In summary, we found that the steady-state error depends on the input type
and the type of system. The table below shows the values of the error
constants and their corresponding error in steady state.

TYPE OF
SYSTEM

ERROR CONSTANTS

ess FOR TYPE OF INPUT

Kp

Kv

Ka

STEP

RAMP

PARABO
LA

1/(1+ Kp)

1/ Kv

1/ Ka

Figure 12-16 TABLE FOR STEADY-STATE ERROR

Take a type 1 system to exemplify the interpretation of the table above. Such a
system is able to follow a step input with zero steady-state error. Now, for the
ramp input, type 1 system is able to go with it in a stationary operation, but
with a finite error given by (1 / Kv). On the other hand, in the case of a parabolic
input, the system type 1 is not able to follow this dynamic and steady-state
error tends to infinity over time.

FREQUENCY RESPONSE
The term "frequency response" refers to the steady state response of a system
with sinusoidal input. So, we abandoned the standard inputs of step, ramp etc
in order to feed the system with a sinusoidal excitation and we can study now
the response behavior of the system under analysis subject to oscillatory
entries. A scan of various input signal frequencies is performed. This type of
approach is very effective because of existing facilities and precision of
measurements. Some Transfer Functions of complicated components are
23

Control

determined using this technique. The methodologies and tolls to accomplished

the analysis of frequency response are, mainly, BODE diagrams, NYQUIST plot
and NICHOLS chart.
BODE DIAGRAMS
If you submit a linear system to a sinusoidal input with constant frequency ,
you will get, after the decay of the transient, a response also sinusoidal with
the same frequency .

Figure 12-17 LINEAR SYSTEM WITH SINUSOIDAL INPUT

The differences between input and output are the amplitude and the phase.
The value of differences varies with the frequency . Bode diagrams record the
variation of magnitude (amplitude) ratio and the phase advance or delay.
Examples of these plots are shown below.

Figure 12-18 BODE DIAGRAMS

From the above diagrams, we can analyze the performance of the set (control
+ system) and evaluate the design to obtain desired specifications or
performance requirements. The main characteristics are:
Bandwidth
Cutoff frequency
24

Control and FBW

Resonance peak
Resonance frequency
Overshoot
Attenuation slope
Gain and Phase margin
Today there are software to produce these diagrams (MATLAB for instance).
Therefore, the design of control systems is much easier.

NYQUIST STABILITY CRITERIA

Consider a closed loop system as in figure 12-15 with a transfer function as

shown below:
Y (s)
G(s)
=
X ( s) 1+G ( s ) H (s)
we saw that the criterion of absolute stability says that to exist stability, all
roots of the equation 1+G(s)H(s) must be in the left half-plane of the plane S.
The NYQUIST STABILITY CRITERION relates the frequency response of open loop
G(j)H(j) to the number of zeros and poles of the equation 1+G(s)H(s) are in
the right half-plane of the plane S. Its advantage is to determine the absolute
stability without being required to calculate the poles of the closed-loop.

25

Control

Figure 12-19 NYQUIST PLOT

NICHOLS CHARTS

A Nichols chart displays the magnitude (in dB) plotted against the phase (in
degrees) of the system response. Nichols charts are useful to analyze openand closed-loop properties of SISO systems, but offer little insight into MIMO
control loops.
An example from MATLAB is displayed below.

26

Control and FBW

Figure 12-20 NICHOLS CHART

ROOT LOCUS

The stability and the basic characteristic of the transient response of a closedloop system can be determined from the position that the closed-loop poles
occupy in the complex frequency plane S.
In the design of closed loop systems, we adjust the parameters in a way that
we can locate the poles and zeros in desired places. The variable that is
available to be modified in a project is, normally, called GAIN of the system
denoted K.
27

Control

Figure 12-21 CLOSED LOOP SYSTEM WITH HIGHLIGHTED GAIN (K)

The method of the locus of the roots was developed by Evans, W. R. and
consists in carrying out a mapping of the poles and zeros from the closed loop
transfer function in the S plane, the extent to which the system gain varies
from 0 to +. The result of this mapping is called ROOT LOCUS of the
system.

Figure 12-22 ROOT LOCUS EXAMPLE

28

Control and FBW

12.5.

CONTROL SYSTEMS APPLICATIONS

The increasing complexity of embedded systems and the improvement of

computational techniques have emphasized the importance of control systems theory.
Topics such as linear control, robust control, adaptive or predictive control have been
target of new developments and design techniques directed to automation of
functions on board, reducing pilot workload and increasing the possibilities and
performance of modern aircraft.

12.5.1. Flight Control Systems

The flight control systems also progressed from mechanical to hydraulic

powered to finally electronic computers as part of the chain of command of
flight. This allowed the introduction of flight monitoring systems, flight
envelope control, and increment of natural stability of the aircraft (eg: SAS =
stability augmentation system). These new features allow a safer flight and
even the performance increase (eg: relaxed stability). On the other hand, these
possibilities have increased the workload of designers and responsibility of
certification authorities. Testing and requirements compliance demonstration
become more complex.

Figure 12-23 GENERALIZED FLIGHT CONTROL SYSTEM

Stability Augmentation Systems are automatic control devices that supplement

a pilot's manipulation of the aircraft controls and are used to modify inherent
aircraft handling qualities. Abbreviated SAS, are essentially damping devices.
They have various forms of auxiliary subsystems added to the primary flight
control to achieve desired aircraft characteristics by selection of variable gain
in feedback loops from aircraft control surfaces. These devices often have
limited authority and do not move the pilot's controls. In many aircraft, an
inoperative SAS makes the aircraft unairworthy.

29

Control

For illustration, a yaw damper is a device used on many aircraft (usually jets
and turboprops) to damp (reduce) the rolling and yawing oscillations known as
the Dutch roll mode. It requires yaw rate sensors and a processor that provides
a signal to an actuator connected to the rudder.

Figure 12-24 SAS EXAMPLES

More information about Flight Control Systems, refer to Chapter 13 GENERAL

SYSTEM.

12.5.2. Fly-by-wire concepts

A fly-by-wire (FBW), by definition, it is a flight control system wherein the

transmission of the control stick signal to the mechanism, which will actuate
the control surface, is performed by a transmitted electrical signal thru a
wiring. However, the fly-by-wire technology, currently implemented in
commercial and typical military aircraft, covers much more than just the
transmission of command through an electrical signal.
In a conventional aircraft, with a mechanical flight control system or powered,
the movement of cockpit controls (control wheel and pedal) are amplified
mechanically or hydraulically and then applied to the various control surfaces.
The force required to operate the surface is supplied directly by the pilot (in the
case of mechanical systems) or amplified by servo-hydraulic mechanisms.
However, the responsibility of coordinating the various movements of the
control surfaces belongs to the pilot.
On the other hand, an aircraft with fly-by-wire system, sensors in the cockpit
mechanisms (stick and pedal) measure the pilot inputs and send them to one
or more electronic modules dedicated to indicate what the aircraft should do.
Computer systems interpret the signals from these and several other system
sensors and generates appropriate commands to the various actuators.
Sensors on the control surfaces continuously feedback their signals to the flight
30

Control and FBW

computers. With this technology, it reduces pilot workload on operation of the

aircraft. Moreover, this method allows the implementation of algorithms which
monitor, process and modify the original pilots request in order to facilitate the
pilotage, such as laws to increase control (Control Augmentation Systems CAS) or increase stability (Stability Augmentation System - SAS). As a result, it
becomes possible to perform a series of improvements in aircraft performance
and design that without the fly-by-wire would not be possible. For example, an
increase in cruise speed, increased maneuverability (crucial for military
aircraft), increased safety margin to prevent exceeding the safety limits during
operation, among other characteristics such as weight reduction and
complexity of flight control mechanisms, etc.

Figure 12-25 FLY-BY-WIRE CONCEPTUAL SCHEME (Source: [Link])

12.5.3. Other applications

AUTOPILOT AND FLIGHT DIRECTOR

An autopilot can be capable of many very times intensive tasks, helping the
pilot focus on the overall status of the aircraft and flight. Good use of an
autopilot helps automate the process of guiding and controlling the aircraft.
Autopilots can automate tasks, such as maintaining an altitude, climbing or
descending to an assigned altitude, turning to and maintaining an assigned
heading, intercepting a course, guiding the aircraft between waypoints that
make up a route programmed into an FMS, and flying a precision or
31

Control

nonprecision approach. You must accurately determine the installed options,

type of installation, and basic and optional functions available in your specific
aircraft.
Many advanced avionics installations really include two different, but
integrated, systems. One is the autopilot system, which is the set of servo
actuators that actually do the control movement and the control circuits to
make the servo actuators move the correct amount for the selected task. The
second is the flight director (FD) component. The FD is the brain of the
autopilot system. Most autopilots can fly straight and level. When there are
additional tasks of finding a selected course (intercepting), changing altitudes,
and tracking navigation sources with cross winds, higher level calculations are
required.
The FD is designed with the computational power to accomplish these tasks
and usually displays the indications to the pilot for guidance as well. Most flight
directors accept data input from the air data computer (ADC), Attitude Heading
Reference System (AHRS), navigation sources, the pilots control panel, and the
autopilot servo feedback, to name some examples. The downside is that you
must program the FD to display what you are to do. If you do not preprogram
the FD in time, or correctly, FD guidance may be inaccurate.
In every instance, you must be sure what modes the FD/autopilot is in and
include that indicator or annunciator in the crosscheck. You must know what
that particular mode in that specific FD/autopilot system is programmed to
accomplish, and what actions will cancel those modes. Due to numerous
available options, two otherwise identical aircraft can have very different
avionics and autopilot functional capabilities

Figure 12-26 AUTOPILOT SCHEME EXAMPLE (SINGLE AXIS)

FADEC ("Fully Authorized Digital Engine Control")

FADEC is a system that controls the overall operation of the engine in response
to a command from the cockpit. Also provide information to the aircraft (flight
deck indication), including information on the engine condition. Another
functions (among others):
32

Control and FBW

You can control the fuel, N1 and N2.

Check the engine parameters during the process of "starting motor" and
prevents the engine Hi EGT that exceeds the limit maximum permissible.
System works in two ways: manual and autothrust.
Optimize the operation of the engine by controlling the flow of air
compressors and the turbines operation parameters.

Figure 12-27 FADEC

AIR CONDITIONING CONTROL Temperature control system.
FLIGHT CONTROL SYSTEM Ride smoothing (turbulence) and Single engine
flight optimization.

12.6.

TESTS AND CERTIFICATION

12.6.1. Ground and Flight tests

The final proof of design of any control system is the demonstration of

successful integration of the control system into the total vehicle of which it is
a part. Such a demonstration of satisfactory and therefore safe design must
include an assessment of the behavior of the vehicle and its control system
over the full range of normal and extreme environmental conditions to which it
will be subject. Thus, in the case of an aircraft flight control system, this can
only be achieved by testing the flight control system (FCS) when it is fully
33

Control

integrated into the aircraft and assessing it over the full operating envelope of
the aircraft. Flight testing therefore represents the ultimate proof that the
design of the FCS is fit for purpose, meets the design requirements and verifies
that the design requirements themselves were valid.
Historically, the flight test process has been viewed as an independent check of
the whole aircraft and its systems by the flight test team of pilots and
engineers whose task was to assess the aircraft behavior and identify any
problems with the design. Any problems identified had to be understood and
resolved by the designers and the modified aircraft reassessed by the flight
test team.
As the system get more complex, the ground testing becomes more
complicated and must embrace all aspects. Based on the ground results, the
flight testing can be programmed.
The key steps are:
off-line design phase;
pilot in the loop simulation;
iron bird tests;
clearance to flight; and
validation thru flight testing.

12.6.2. Certification aspects

How did the FAA certify Part 25 Fly-By-Wire (FBW)?

Current Rules do not address modern FBW
FAA did not dictate design
FAA accepted Boeing + Airbus approach

But FAA levied Special Conditions

Fit into existing regulatory structure

12.7.

REFERENCES

OGATA, Katsuhiko Engenharia de Controle Moderno 3 Edio

34

Control and FBW

DAZZO, J.D. & HOUPIS, C.H. Anlise e Projeto de Sistemas de Controle Lineares 1979
(Anonimous) Feedback Control Theory USAFTPS - 1982
GUIMARES NETO, Ary Apostila do Curso de Ensaios em Vo CEV 1986

([Link]
%20Lowell%[Link] accessed 10 FEB 2016)

35

Control

APPENDIX 12A
EXERCISES
1 - Explain the concepts:

Block diagram
Steady-state performance
Transient phase parameters performance
Time domain and frequency domain
Transfer function

2 - Find the Transfer Function of the system below:

3 - What kind of information comes from root-locus and bode diagram that
helps the control system analysis and design?

4 - Prove that the relation below gives the multiplication of complex numbers:
(a + bj) . (c + dj) = (ac-bd) + (bc+ad)j

5 - Sketch the step response of a system of first order and another of second
order. Show the main performance parameters on the sketches.

36