Seminar Report Final
Seminar Report Final
AIRCRAFT ASSEMBLY
A SEMINAR REPORT
Submitted by
ANANDU ARAVIND
(SBC17ME041)
to
of
Bachelor of Technology
In
MECHANICAL ENGINEERING
NOVEMBER 2020
DEPARTMENT OF MECHANICAL ENGINEERING
CERTIFICATE
First of all, I am grateful to The Almighty God for establishing me to complete this
seminar. I wish to express my sincere thanks to Dr. S. Suresh Babu, Principal of the
college, for providing me with all necessary facilities and support.
I also thank my Parents and siblings for their unceasing support and encouragement.
I wish to express my sincere thanks to all my friends for their goodwill and
constructive ideas. I also place on record, my sense of gratitude to one and all who,
directly or indirectly, have lent their helping hands in this venture.
ANANDU ARAVIND
S7 MECHANICAL
i
ABSTRACT
ii
LIST OF CONTENTS
CHAPTER
TITLE PAGE NO.
NO.
ACKNOWLEDGEMENT i
ABSTRACT ii
LIST OF FIGURES v
ABBREVIATIONS vi
NOTATIONS vii
1 INTRODUCTION 1
3 SEMI-AUTOMATED APPROACH 4
4 FREQUENCY IDENTIFICATION 9
iii
INDEPENDENT AMPLITUDE
5 VALIDATION 11
6 CONCLUSION 12
REFERENCES 13
iv
LIST OF FIGURES
FIGURE PAGE
TITLE
NO. NO.
v
ABBREVIATIONS
vi
NOTATION
f FREQUENCY, Hz
m MASS, kg
vii
CHAPTER 1
INTRODUCTION
Project objective was the fast and effective set-up of individual HRC robot systems,
which are customizable regarding the needs and requirements of the scenario. The
modular approach of the developed components is not only limited to the hardware of
the robot, but is also transferred to the implemented software. Apart from the HRC
robot arm design, video projectors are used to visualize work spaces and monitor
them in combination with camera systems to complete the safety of the whole HRC.
With respect to the usability of the robot system, a modular control concept is used, as
well as multimodal interaction, like voice and gesture control, and hand guiding or
positioning of the robot system using a touch screen. Enhancing the effectiveness is
achieved due to the applied HRC in the applications as an easy to maintain and
reconfigurable modular robot system.
In order to validate the results of the project, application scenarios such as,
manipulation, bolting, welding, and riveting were examined within the Four By Three
project.
1
CHAPTER 2
The aircraft is subdivided into several sections. The cockpit is located in the first
sections, while sections 13-18 from the fuselage, which represents the passenger area.
The aft section of the aircraft, section 19, closes the passenger area with the pressure
bulkhead and the section barrel. In addition, the horizontal and vertical tail planes are
mounted to section 19. Within the scope of assembly system, the section 19 assembly
of an Airbus A320 is examined more closely. As shown and marked red in Fig 2.1,
the pressure bulkhead needs to be mounted to the section barrel 19. This assembly
process is currently performed in box assembly by means of a high number of rivets.
The overlapping riveting position of the pressure bulkhead with the section barrel is
marked orange in Figure 2.1.
The size of the components interferes with the tight tolerances of manufacturing and
assembly processes. Especially when it comes to automated systems for aircraft
production is often difficult and expensive to meet the requirements and to identify
the different variations which are introduced by a large number of parts and complex
joining tasks. As a result, many assembly processes in structural assembly, such as
the application example of the mounting of the pressure bulkhead to the section
barrel, are still not automated. In this assembly process many burdensome and tiring
task, such as monotonous overhead work, high noise level due to the riveting hammer
and bad ergonomics related to the body postures have been identified within the
2
process analysis. Two operators (Fig 2.2) must work manually and cooperatively as
the pressure bulkhead is mounted to the barrel. Preparation tasks include the pre-
drilling, drilling out, sinking, disassembly, and the application of sealants to
approximately 800 riveting points. The first operator kneels on top of the pressure
bulkhead, identifies the riveting position, inserts the rivets, and operates the riveting
hammer. The second operator needs to be inside the section barrel to position the
anvil for the collaborative riveting process. To reach the work space, the operators
require racks on the outside and lifting units on the inside of the section. In addition
to the troublesome overhead work required by the operator performing the counter-
holding task, the exposure to noise and vibrations by the tool is amplified while
inside the section barrel. The operator inside the section barrel must therefore remain
in an ergonomically unfavourable position, in a confined and noisy workspace, for
about 3 to 4 hours of process time. Furthermore the noise makes communication
between the two operators more difficult and only possible by means of knocking
signals, which decreases the process efficiency.
3
CHAPTER 3
SEMI-AUTOMATED APPROACH
The planning approach used is shown in Figure 3.1, and the subsequent detailed
analysis highlight the importance of understanding the dependencies between these
important components of a production process. The resulting requirements obtained
from such an analysis were incorporated into the process design and the production
equipment selection.
Through skill-based task allocation, the human was placed on top of the pressure
bulkhead to perform the more complex task of inserting the rivets and operating the
riveting hammer in the joining process. The robot was placed inside the barrel to
position the anvil, as it is a monotonous and non-ergonomic tasks. This collaborative
method thus results in semi automation of the production process which is controlled
by a higher level control system. The overlapping of work spaces between the human
and the robot system at the same time and the collaborative work on a common task,
such as the riveting production process, defines the configuration as Human-Robot-
Collaboration. In terms of safety, the removal of barriers leads to increased safety
requirements for the system. The robot system and its lifting unit must conform to
safety standards such as DIN ISO/TS 15066 to ensure that the operator is not harmed.
While market ready robot systems may already fulfil these safety requirements, the
5
lifting unit needs to be outfitted with additional sensors, safety switches or contact-
free sensors. Within this scenario a Human-Robot-Collaboration capable robot system
as well as a laser scanner for monitoring of the work space guarantees safety. While
in the riveting process human and robot are only connected by means of the rivet
itself, separation of both participants by the pressure bulkhead avoids collisions with
the process execution and increases therefore safety.
The robot controller itself is not sufficient, to properly manage the whole system a
higher level control system is needed. The robotic framework Robot Operating
System (ROS), which uses a service-oriented architecture, can serve as a base for the
development of such a system. ROS provides a collection of tools and libraries and
can make robot applications available on a variety of platforms. It can be used for
communication between machines in general, robots, sensors, and additional devices.
For instance, additional sub-programs have been developed as well as graphical
interfaces or the implementation of smart devices. The modular architecture and
standardized communication protocols such as ROS or MQTT publish and subscribe
methods allow an easily transferred to many robotic platforms without significant
effort. For increased efficiency and accuracy in the production process, an adapted
measurement concept is necessary to precisely determine the rivet’s position. The
identification of the work-piece or the riveting pattern is achieved through the use of
an optical sensor. Rivet holes can be detected by the sensor during a scan process and
the position of the rivet holes with respect to the robot base (x, y, z, α, β, γ) can be
determined for the creation of the work model. Additionally, the scanned data can be
combined with offline data based on a CAD model to make use of hybrid
6
programming. This makes it possible to put the application into operation quickly and
increases the efficiency of the application. In terms of process-specific tools, for this
riveting application, the robot is equipped with an anvil for the counter-holder task.
The cold forming process, which results from the interaction of the riveting hammer
and the anvil, requires a special riveting tool. This tool is needed since the riveting
process of a universal rivet requires vibrations that must be minimized to avoid
damages to the robot system.
In aircraft assembly many different types of rivets have various roles in joining
processes, such as pop, hollow, punch or solid rivets. They are additionally
distinguishable by head shape, such as countersunk, round, flat, or universal head. In
this application, a solid rivet with a universal head is required. The HRC riveting
process therefore requires the development of a specific counter-holder tool for the
robot system. Requirements for the tool include several aspects. First the tool has to
be modular for regular maintenance of important parts such as the anvil mass due to
erosion. A safe HRC design can be guaranteed by means of rounded edges and
limitations of gap width. Additionally the design has to be adjusted with respect to
work space, sensors, such as a force-torque sensor for HRC interaction needs to be
integrated, laser line triangulation sensors for process control and referencing, and
damping of vibrations caused by the riveting process. One of the key features of the
process tool is the minimization of process forces and vibrations, which are applied
by the riveting hammer into the rivet and therefore also into the robot system.
Applying the forces, especially the peaks and permanent oscillations, to the robot
could result in damage to parts of the system, such as, the joints and gears. To design
the parameters of the process tool in such a way that minimizes the process forces, the
7
natural frequency of the robot system and the anvil needs to be identified for the
modelling of a dual-mass system. Afterwards optimized parameters for the tool can
be calculated and tool prototypes can be built (Figure 3.4).
8
CHAPTER 4
FREQUENCY IDENTIFICATION
To apply the dual-mass system the parameters of the robot system and the anvil must
first be determined. Experimental modal analysis by means of a laser tracker is used
to determine the stiffness c1. The robot flange is loaded with 2.5 kg while the wrist is
oriented upwards. The displacement of the Tool Centre Point (TCP) is monitored
with a laser tracker (Type: API Radian R50) and measured as 0.475 mm in the
direction of earth gravity. The stiffness c1 is then calculated as 52 N/mm. The natural
frequency of the UR 10 robot must be determined. For this purpose, the robot system
is stimulated to vibration and the system response is measured. The stimulation takes
place by means of a hammer and dynamically by abrupt stopping of the robot during
its planned movement execution. The signal is measured at 100 Hz with respect to the
Nyquist rate to avoid measurement errors. The measurement data is displayed as a
function of time (Figure 4.1(a)) and therefore a discrete signal. By using a discrete
Fourier Transformation the discrete signal can be converted into the frequency range
(Figure 4.1(b)).
(a) (b)
The corresponding frequency f1 is determined as 9.4 Hz. The effective mass of the
9
The mass of the anvil m2 is known as 2 kg. The anvil’s spring stiffness c2 is
calculated similarly to that of the robot. Where the anvil is loaded with 3.5 kg and the
displacement is monitored. Then, the stiffness c2 is calculated as 11 N/mm.
By using approximate equations and solutions the natural frequencies of the duel-
mass system are determined as w1 = 8 Hz and w2 =13 Hz.
Using the previously calculated values, as well as the two-mass system stimulated by
the riveting hammer with a harmonic oscillating amplitude (f = 29 Hz), values are
determined and plotted as shown in Figure 4.2. The plot in Figure 4.2 shows the
amplitudes of the robot and the anvil, as well as the effacement point. The plot shows
two characteristics. Firstly, the areas of resonance for the robot and the anvil are in
overlapping areas. Secondly, the effacement point that would cause the anvil to stop
swinging is approximately at 10 Hz. The riveting hammer frequency has been
determined by means of an acoustic level measurement. The riveting hammer
frequency allows the operation of the riveting process in a safe frequency domain that
is far from the resonance area of the two-mass system. As a result, damage to the
robot manipulator especially into gears during the production process is avoided
(Figure 4.2)
10
CHAPTER 5
VALIDATION
For validation of the developed modular riveting process tool, the demonstrator,
which was developed within the framework of the Four By Three project, is used.
The previously shown and determined values for the design of the process tool were
implemented into a real prototype tool in order to validate the results in practice. To
validate the oscillation insulation by the tool, the forces applied to the robot flange
during the riveting process are measured. In order to be able to make a statement on
how much the spring system reduces the forces and oscillation, a reference
measurement was made. The mean value of the forces for a riveting process without a
spring is 125 N. In addition, peak forces of up to approximately 195 N occur, which
would cause damage to the robot if they were permanently exposed. By using the
developed riveting tool with adapted parameters for minimizing forces and
oscillations, finally averaged forces of approximately 65 N are measured. In practical
comparison, the forces can thus be minimized by about 50%.
11
CHAPTER 6
CONCLUSION
12
REFERENCES
13