Control science and automation

hourglassjurorΜηχανική

5 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

323 εμφανίσεις

 
Control science and automation
L. Brunetti

A special skill of LAPP is automation and control science. This activity is dealt by a group
which belongs to the mechanical department and is composed of three permanent employees
(one research engineer who is supervising this group and two assistant engineers) and
occasional short term collaborators.


Automation:


The activities dedicated to the automation has began several years ago with different projects,
particularly the development of the drive system of the LHCb detector and the automatized
loading / unloading system of the 10 000 bricks of the Opera detectors. This activity still
under development. Currently LAPP is in charge of the realization of the full automatized
loading / unloading system of the camera of the HESS experiment, which is the biggest
particle physics telescope in the world; this type of operation has never been done before.
Given these past and current important developments, the LAPP is recognized to be one of the
few laboratories of the IN2P3 which is able to perform large size and complex automatized
processes. This achievement has been made possible because of on one hand the synergy of
different domains of competences (like instrumentation, mechanics, electrotechnicals and
computer science skills) in the laboratory and on the other hand the collaboration with the
different specialist entities of mecatronics in the region.
Through these experiments, the LAPP is close to the automation industrials fields, i.e. the
local manufacturers like cabling or electrotechnicals suppliers, or to the specific
manufacturers dedicated to the automation or computer science.
LAPP has set as its target to develop and deepen this skill in order to carry out the future
projects like CTA, for which we have planned to carry out the drive system of the 4 future
large telescopes.
Furthermore, considering that a lot of current issues of automation are the management of
deported multi-process control of large dimensions. This requires to carry out networks of
PLC (Programmable Logic Controller) with deported components (sensors, actuators,
motors). These components are linked through deterministic fieldbus in order to respect the
current increase of the data flow on continually growing distances. The aim is to become one
of the leaders in the PLC architecture definition integrating the optimal industrial fieldbus for
the CTA project, knowing that this subject is also covered by one of the LPSC projects
concerning the cooling. We should also mention our local industrial partnerships which have
the same type of problems with various complex automated industrial systems, such as
manufacturing assembly line.
In this prospect, the LAPP has to improve the dedicated installation and equipment.
Indeed, LAPP plans to arrange a dedicated room with handling tools and specific equipments
in order to be able to do the electrotechnicals part of the projects in better conditions. In this
room, we would like to carry out a sophisticated test bench which will allow us to develop
various drive system configurations dedicated to the current and future experiments.
Furthermore, this test bench will fit with our developments in the fieldbus aspect with
networks of PLC and with different architectures that we would like to carry out and to
compare.
 
This new “automation platform” will be usable by our partners, especially CETIM and
SYMME. It requires also human resources in order to start up all the architecture, the training
of the actual control science team and some new software.

Positioning:


The control science group has to support the needs of the different experiments in this specific
field.
In the future, different projects in particles (ILC, ATLAS) or astro-particles (SiPM, PMT) will
require a positioning system for different applications. The wish of LAPP is to invest in
different standard industrial products in order to answer the needs of the different experiments
and have a set of positioning systems which can be used by our partners or eventually rent to
the local companies.

The first identified need is a positioning table with the main characteristics which are detailed
in the table 01. This table will be used for the beam tests of the future detector of ILC and will
be located at CERN. Actually, there are not available positioning tables with the desired size
and precision at CERN and this equipment will be probably used for other laboratories at
CERN.

Experiment ILC
Load 10 tones
Objet to be positioned Detector
Translation axis Tx, Ty
Range in translation 1 200 x 1 200 mm
Accuracy (in translation) 1 mm
Speed 10 mm/s
Rotation axis Ry
Range in rotation 0 - 90°
Accuracy (rotation) 1°
Command Deported manual command
Interface PC via Ethernet (OPC server)
Table 01: needs for the large positioning table

A second need, which is very common for different projects, is the possibility to place an
object relatively to a target or to scan an area. The main characteristics are detailed in the
table 02 and product of the society Newport were selected [aa].

ILC ATLAS SiPM PMT
Quantity 1 1 2 2
Load (Kg) 10 0.5 3 5
Object Radioactive
source
Laser SiPM PM
Environment Grey room White room Black room Black room
Translation axis Tx, Ty Tx, Ty Tx, Ty Tx, Ty
Range in translation 350 x 500 mm 50 x 50 mm 150 x 150 mm 320 x 320 mm
Accuracy (translation) 0,5 mm 1 µm 0.01 mm 0.1 mm
Speed 10 mm/s 5 mm / s 5 mm / s 5 mm / s
Rotation axis - Rx, Ry - -
Range in rotation - +/- 45° - -
Accuracy (rotation) - 0.1° -
 
Command Automatized
cycle
Automatized
cycle
Deported
manual
command
Deported
manual
command
Interface PC via Ethernet
(OPC server)
PC via Ethernet
(OPC server)
PC via Ethernet PC via Ethernet
Table 02: needs for the small positioning table or scanning

Control science:


LAPP, in partnership with SYMME, CETIM and in collaboration with CERN, has been
involved in the vibration control for the future linear collider CLIC for several years.
Different aspects are under study. The first one is a feasibility demonstration of the capability
to control a mechanical structure at a sub-nanometer scale [bb].
At this moment, the group is focused on the development of active isolation in order to
decrease the motion of mechanical structures at the sub-nanometer scale (i.e. focusing
magnets). There are industrial products [cc], but the specifications are not completely
respected and they are very expensive. Various studies were performed [dd] but none was
sufficiently close to the requirements. As a consequence, our purpose is to develop a low cost
dedicated table, knowing that this future product would be usable for various applications in
laboratory and industrial field like optics.
The group works also on the beam trajectory control through the two last focusing magnets
where the motion of the beam has to be as low as possible (0,1 nm). The current study has
provided encouraging results in simulation and is still in progress.
This topic is still under development at LAPP, since we have recently became in charge of the
mechanical stabilization of the arches of the four future large size telescope of CTA.
In this context, LAPP has to be able to measure ground motion and mechanical structures
motion with specific and advanced instrumentation combined with very efficient real time
solution in order to perform the dedicated control laws which are developed.
Currently LAPP is equipped with a real time solution [ee] (which is not adapted anymore) and
a limited set of sensors actuators. As a result, the target is to replace the existing equipment
with a new generation of real-time solution [ff] and to purchase two seismic sensors
(geophone Guralp) and two accelerometers (Endevco or Willcoxon).
Thanks to this equipment, LAPP will be able to reinforce their activities in vibrations control
and will be able to support more often various particle physics experiments in seismic motion
measurement campaigns as it was already done for CERN or SuperB at Frascati.
Note that one of the current limitations is the quality of the manufactured sensors (accuracy,
internal noise, size, sensitivity to radiation). Next, given the acquired experience in this field,
it is possible for the development of a specific sensor to start in the future in partnership with
manufacturers.

Robotics:


In the continuity of the current activities of the control science group, LAPP would like to
extend its application domain by developing the robotics. This field is strategic for the control
science at LAPP, but it also in phase with the future activities of our partners. This
implication will be done in synergy with them inside a common building which is being built
called the “maison de la mecatronique”.
In this prospect, two applications (ILC & ATLAS) require developments in robotics.

• Automatized tests of chips:
 

The future ILC detector is composed of many chips which have to be tested. For the next
stage, 5000 chips have to be tested and this operation cannot be done manually. Thus it would
be necessary to carry out an automatized test bench including an industrial 4 axes robot [gg].
It will be also the opportunity to purchase new equipment which could be used for other chips
test campaigns but also for various application of tests, assembling, manipulation…
It will require also some training of the control science group.

• Computer Assisted Teleoperation:

Maintenance of detectors in irradiated area has become very complex with the growing
energy accelerators such as LHC at CERN. As a human being cannot spend a long time in
such environment, robotic systems must compensate human actions.
This problem is common with other very specific fields [hh] where the Computer Assisted
Teleoperation (CAT) is developed but it has never been done in the particles physics. As a
result, LAPP would like to be the first one in developing teleoperation assisted by computer
dedicated to particle physics.

The equipment consists of a 6 axis industrial robot, which has to be instrumented in order to
obtain a force feedback solution. It is also interfaced with software of computer assisted
conception using a specific software module in order to simulate the trajectories of the robot
in 3 dimensions in a safe environment. Given the planned handling operations which have to
be performed, the laboratory will be equipped with two such devices.

At this moment, the ATLAS experiment would perform the disconnection of facilities with
this solution, so the robot will be able to work in a cylinder of 2 m of diameter and of 10 m
long (a part of the detector), as well as outside this cylinder. However, other experiments may
be interested on this process in the future.

Note that all the details of the planned investments are detailed in the specific xls file
document.


Bibliography

[aa] Newport corporation: http://www.newport.com/

[bb] B. Bolzon, “Etude des vibrations et de la stabilisation à l'échelle sous-nanométrique des
doublets finaux d'un collisionneur linéaire”, Thèse, Université de Savoie, 2007.
[cc] TMC Company: http://www.techmfg.com/index.html

[dd] C. Montag, “Active stabilization of mechanical quadrupole vibrations for linear
colliders”, Nuclear Instruments and Methods in Physics Research, A 378 (1996) 369-375.
[ee] XPC Target (real time solution) of the society Mathworks:
http://www.mathworks.com/products/xpctarget/

[ff] DSpace real time solution: http://www.dspace.com/ww/fr/fra/home.cfm

[ii] B. Bolzon, L. Brunetti, A. Jeremie, S. Tomassini, U. Rotundo, M. Esposito, “Preliminary
ground motion measurements at LNF site for the Super B project”, IPAC 2010, Kyoto.
[gg] STAUBLI company: http://www.staubli.com/fr/robotique/

[hh] P. Desbats, F. Geffard, G. Piolain, A. Coudray, “Force-feedback teleoperation of an
industrial robot in a nuclear spent fuel reprocessing plant”, Industrial robot, 2006, vol. 33