Research area 1 - SFI Norman

sillysepiaElectronics - Devices

Nov 27, 2013 (3 years and 6 months ago)

75 views

sfinorman.no


1

WP1


Robust and Adaptive Manufacturing Systems

WP2
-

Advanced Process Control
and Intelligent Maintenance

WP3
-

Hybrid Manufacturing

GOAL:

Develop system concepts for automated manufacturing with high performance based on integration and
adaptivity

in manufacturing systems

GOAL:

Develop knowledge, tools, and concepts for
advanced process control and intelligent
predictive maintenance of equipment for high
performance
manufacturing

GOAL
:

Develop the concept and principles for a hybrid
manufacturing system



RA1:
Advanced Manufacturing Technology


sfinorman.no


2

WP1
-

Robust
and Adaptive
Manufacturing Systems

WP2
-

Advanced
Process

Control

and

Intelligent
Maintenance

WP3
-

Hybrid
Manufacturing

WP6

Research area 1:

Advanced Manufacturing Technology

T3

WP4

Planning and
Control

WP5

Work
Organization

T4

T2

T5

Collaboration between WPs

sfinorman.no


3

WP1

Robust and Adaptive Manufacturing Systems

Implications of the

concept of the
constantly changing manufacturing system

f
or:

T1:

Study new design methods for manufacturing control based on an agent
-
oriented
bottom
-
up approach

T2:

Develop and integrate new agent
-
oriented design tools in the APROX framework

T3:

Define operator information and control requirements in highly automated
manufacturing environments
-

work
organization

and
demand for skill development

T4:

Define
handling characteristics for non
-
rigid materials

WP2

Advanced Process Control and Predictive Maintenance

T1:

Sensor and sensor system development and integration

for measurement of critical
process parameters

T2:

Control
strategies and
methods

for
self
-
adjusting,
-
calibrating and
-
reconfigurable
processes

T3:

Fault

diagnosis

and

prognosis

system

for

preventive

m
aintenance

of

production

equipment

T4:

3D
-
object m
easurement and inspection
on
the basis of 3D point
clouds

T5:

Operator decision
-
support:
strategies
,

models and tools for effective problem solving based
on a combination of
operator/specialist
knowledge and
monitoring of measured or estimated
process parameters

WP3

Hybrid
manufacturing

T1:

Development of a
hybrid
m
anufacturing
cell
by i
ntegration
of

a
dditive
m
anufacturing with
conventional CNC
milling

T2:

Case studies:
p
rinciples
for enhanced tooling capability
and
high performance parts by
incorporation of complex
geometries
and variable material
composition
for advanced thermal
management and directed part material properties

T3:

Design for performance:
design
principles
to
exploit the possibilities of the Hybrid
Manufacturing
concept

Task in all WP's: International
collaboration and network
building

PhD involvement

sfinorman.no


4

WP1

Robust and Adaptive Manufacturing Systems

Implications of the

concept of the
constantly changing manufacturing system

f
or:

T1:

New design methods:

symbolic communication between machines/devices. For such
communication, both software and hardware of present equipment must be extended.

T2:

Develop and integrate new agent
-
oriented design tools:
systems, e.g. assembly systems,
capable to work in not well structured environment.

T3:

-

work
organization

and
demand for skill development

T4:

Define
handling characteristics for non
-
rigid materials

WP2

Advanced Process Control and Predictive Maintenance

T1:

Sensor and sensor system development and integration

for measurement
of
critical
process
parameters:

Sensor n
etworks capable of acquiring symbolic data

T2:

Control
strategies and
methods

for
self
-
adjusting,
-
calibrating and
-
reconfigurable
processes:
strategies and methods based on symbolic data mining and optimization.
Solutions imitating biological reflexes

T3:

Fault

diagnosis

and

prognosis

system

for

preventive

m
aintenance

of

production

equipment

T4:

3D
-
object m
easurement and inspection
on
the basis of 3D point
clouds

T5:

Operator decision
-
support:
strategies
,

models and tools for effective problem solving based
on a combination of
operator/specialist
knowledge and
monitoring of measured or estimated
process parameters:
HMI communicating with operators on the symbolic level

WP3

Hybrid
manufacturing

T1:

Development of a
hybrid
m
anufacturing
cell
by i
ntegration
of

a
dditive
m
anufacturing with
conventional CNC
milling

T2:

Case studies:
p
rinciples
for enhanced tooling capability
and
high performance parts by
incorporation of complex
geometries
and variable material
composition
for advanced thermal
management and directed part material properties

T3:

Design for performance:
design
principles
to
exploit the possibilities of the Hybrid
Manufacturing
concept

PhD involvement

sfinorman.no


5

5

Control logic verification

Before

Programming logic

in QUEST
*

syntax

'Verified' control logic

Programming logic

in target language
**

syntax

Truly verified control logic

in real equipment environment

*QUEST simulation software

**Python

Now

Results from RA1

WP1 Robust and Adaptive Manufacturing Systems

sfinorman.no


6

Control logic verification

Now

Programming logic

in target language
**

syntax

Truly verified control logic in

emulated
equipment environment

Switching to real

equipment environment

Results from RA1

WP1 Robust and Adaptive Manufacturing Systems

sfinorman.no


7

1.
Flexible, automated sewing


further
developed:

+
A
software has been developed for integration of control of robot,
PyMoCo

and
ROS

+
Real
time control

has been tested and promising results have been
achieved for 8 milliseconds
control.

+
A new
speed sensor

(mechanics and electronics) has been developed.
The sensor will be used for measurements required for further
development of the control system for the sewing
cell.

=
Sew together parts of different shapes and materials, without prior
knowledge of the part geometries

Results from RA 1

WP2
Advanced Process Control and Predictive Maintenance




2.
A
predictive maintenance model

has been established in order to
obtain optimal maintenance scheduling based on the condition of
the equipment.


3.
RFID techniques in condition monitoring

has been
researched,
and a demo
of RFID
application
in production system has been
established.


4.
A dual arm robot installation is being built





sfinorman.no


8

1.
A
new method for preparing the substrates for
additive manufacturing

in a CNC milling machine has
been developed
.


2.
The
cohesion of the AM section to the base part has been
tested with excellent
results
(
Marlok

C1650+

CL 50WS AM tool
steel
)
.



3.
Porous
sections built into the tool insert
derived
as a
valuable complement to other practical
solutions


4.
A prototype integrated control system for the hybrid
cell (OMOS)
has been further developed, in collaboration
with exchange student from Slovenia
.


5.
A prototype of the hybrid cell control system has been
developed.



Results from RA 1

WP3 Hybrid Manufacturing



sfinorman.no


9

Other

results

New projects:


Autoflex

-

Flexible automated manufacturing of large and complex
products:
Partners: Rolls
-
Royce
Marine
AS,
Benteler

Aluminium

Systems Norway AS,
Intek

Engineering AS, SINTEF
Raufoss

Manufacturing AS and NTNU
.


SmartTools
: Partners: Sandvik
Teeness

AS,
SINTEF ICT, SINTEF
Raufoss

Manufacturing
and NTNU
IPK



Contribution to education:


The
Framework of IFDPS becomes a part of a course (TPK 4155 Applied Computational
Intelligence in Intelligent Manufacturing)


The RFID application demo for Production System becomes a practice study for a course
called PK8106 Knowledge Discovery and Data
Mining


sfinorman.no


10

International
collaboration

within

RA1 in
2012:

Chairman from Industry for Joining Sub
-
Platform
: SFI Norman and SINTEF
Raufoss

Manufacturing AS
have worked actively in
Manufuture

by participating in the
HLG.
As a result
Kristian

Martinsen

now holds
the chair, as an industry representative, for the new sub
-
platform for Joining.


Exchange
agreement with
four students
from
Ensiame

Engineering School,
Valenciennes
, France.
Have
been working on design of a flexible jig for assembly of components for Sandvik
Teeness

and a dual arm
robot installation.


Collaboration
through the
development of the new ISO standard on additive manufacturing technology

does now include
the chair for ISO/TC261 WG1 Terminology for additive manufacturing.


DTI
(Denmark), VTT (Finland),
Acreo

(Sweden),
Fraunhofer

(Germany):
collaboration on coatings,
integrated sensors and new business models for injection molding industry
.


Two
new EU
-
projects have been granted
, SASAM and
Diginova
, where SINTEF
Raufoss

Manufacturing is
a partner.
Diginova
, short for Innovation for Digital Fabrication, is a coordination and support action
project under NMP 7th FP, Networking of materials laboratories and innovation. SASAM, which is short
for Support Action for
Standardisation

in Additive Manufacturing, is a similar type of project
.


Collaboration on a EU
-
proposal "VITAMIN"
, where Sandvik
Teeness

was partner together with SRM and
SINTEF ICT from Norway. Not granted.

sfinorman.no


11

Planned

international

collaboration

within

RA1 for 2013:

Polytechnic

Institute

of

Braganca
,
Portugal:


Prof
. Paulo
Leitaõ
: workshop
around

holonic

manufacturing
,
common

publication

or
similar
.




The
University of Manchester,
UK
: Dr. Yi Wang:


establishing projects
on
Intelligent systems and Predictive Maintenance.


Common publication: a book on data mining for zero
-
defect manufacturing


VTT Technical Research Centre of Finland,
+rest of consortium


EU
proposal for call FoF.NMP.2013
-
7 "New hybrid production systems in advanced
factory environments based on new human
-
robot interactive cooperation
":


University
of
Ljubljana:


Prof
.
Slavko

Dolinsek

and student David
Homar
,
continue collaboration
on development on
OMOS (Optimized Manufacturing Operation Sequence
)


University

of

Berlin (????

):


Prof
. Günther
Seliger
: workshop
around

flexible

automation

and
possibly

researcher

exchange
?



sfinorman.no


12

More
detailed

on

results

sfinorman.no


13


Flat milling produces a glossy surface;


Low
-
friction for powder spreading


Reflective to laser beam


Standard procedure: Sand blasting,
-
unsuitable for the
hybrid cell


Hybrid cell

procedure: Extra sharp cutting tool inserts
"scratch" the substrate


Provides an exact z = 0
-
point for starting the AM building

Some results from RA1

Substrate preparation


Edge radius: 0


0.1 mm;


Cutting depth: 0.1 mm;


Feed rate: 0.05 mm/O

sfinorman.no


14

1
4

OMOS:

Optimized Manufacturing Operation Sequence



sfinorman.no


15

Results:


Cooling time for conventional insert and “old” design 70 sec.


Estimated cooling time with new design approximately +25 sec. = 95
sec.


Cooling time with new design and conformal cooling insert: 48 sec.


Cost of machining AM produced insert similar to conventional
production, however the cost of AM makes this an expensive insert

Industrial need: reduced cost of production by AM closer to final shape

Some results from RA1


WP3: Industrial case studies: insert for a bracket to an office chair

sfinorman.no


16

Working principle:

Demonstrator development


sfinorman.no


17



Equipment or Process
Degradation Process
Sensors
(
Data Acquisition
)
Feature Extraction
Fault Diagnosis
Fault Prognosis
Maintenance Scheduling
/
Maintenance Optimization
Signal Pre
-
process
Denosing
Time Domain
Time
-
Frequency
Domain
Frequency Domain
(
FFT
,
DFT
)
Wavelet Domain
(
WT
,
WPT
)
Principal Component
Analysis
(
PCA
)
Compression
Extract Weak Signal
Filter
Amplification
Support Support Machine
(
SVM
)
Data Mining
(
Decision Tree
&
Association rules
)
Artificial Neural Network
(
SOM
&
SBP
)
Statistical Maching
Auto
-
regressive Moving
Averaging
(
ARMA
)
Fuzzy Logic Prediction
ANN Prediction
Match Matrix Prediction
Ant Colony Optimization
(
ACO
)
Particle Swarm Optimization
(
PSO
)
Gentic Algorithms
(
GA
)
Meta
-
Heuristic approaches
Bee
Colony Algorithms
(
BCA
)
Information Delivery
Demonstrator development


Example
:

System Frame of
IFDPS


Intelligent Fault Diagnosis and Prognosis System