Slab surface inspection and rectification

EUROPEAN-COMMISSION - Directorate-General For Translation European Commission

Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres

128 pages

English

Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres

A propos
Informations
Extrait

Description

Industrial research and development

Informations

Publié par	EUROPEAN-COMMISSION
Nombre de lectures	14
Langue	English
Poids de l'ouvrage	5 Mo

Extrait

Commission of the European Communities
steel research
Measurements and analysis
SLAB SURFACE INSPECTION
AND RECTIFICATION
Report
EUR 10139 EN
Blow-up from microfiche original Commission of the European Communities
■ mi ι
Measurements and analysis
SLAB SURFACE INSPECTION
AND RECTIFICATION
M.J. HAGUE, M.G. BROOKS
BRITISH STEEL CORPORATION
9, Albert Embankment
GB-LONDON SE1 7SN
Contract No 7210-GA/605
(1.3.1981 - 31.8.1984)
FINAL REPORT
Directorate-General
Science, Research and Development
1986 EUR 10139 EN Published by the
COMMISSION OF THE EUROPEAN COMMUNITIES
Directorate-General
Information Market and Innovation
L-2920 LUXEMBOURG
LEGAL NOTICE
Neither the Commission of the European Communities nor any person acting
on behalf of then is responsible for the use which might be made of
the following information
>ECSC-EEC-Euratom, Brussels- Luxembourg Slab Surface Inspection and Rectification
FINAL REPORT
Agreement 7210.GA/605
M.J. Hague
M.G. Brooks
British Steel Corporation
Teesside Laboratories
EUR 10139 EN FR 107-9 842 7210.GA/605
SLAB SURFACE INSPECTION AND RECTIFICATION -
SUB CONTRACT BY AGREEMENT WITH HOOGOVENS
British Steel Corporation
ECSC Contract No.5
Final Summary Report
The objectives at the start of this project were to firstly prove that optical
inspection systems based on video or charge coupled device (CCD) cameras were
capable of detecting surface defects in both hot rolled slabs and continuously
cast slabs. Following this, an automatic rectification system was to be developed
by which means the identified defect on a continuously cast slab could be removed.
Three courses of action were followed. At the BSC Welsh Laboratories, Port
Talbot, effort was concentrated into the application of CCD cameras to identify
and recognise the most common types of rolled slab defects.
At the BSC Teesside Laboratories, both CCD and video cameras were used to image
continuously cast slabs, and in parallel to this activity the development of a
robotised lorry loader used to carry a scarfing torch for defect rectification was
pursued.
Extensive laboratory tests of camera performance were followed by trials at Port
Talbot slabbing mill and Lackenby (South Teesside) continuous casting plant.
Although a variety of rolled slab defects could be identified, the continuing
trend towards the continuous caster production route in BSC and elsewhere
necessitated most effort being spent in this area. It was considered that an
inspection system for longitudinal defects on the broad face of the slab should be
able to detect a defect of length at least 100 mm and width of at least 1 mm.
The need for descaling the slab surface was paramount and suitable equipment was
hired in order to descale a 400 mm strip along the slab surface by means of high
pressure water jetting. The descaling activity gave rise to problems with the
slab lifting and it was recognised that ideally a higher pressure, lower water
flow rate combination with different spray nozzles and higher pressure pump would
have been preferable. However, the problem was controlled so that useful defect
data from the slab surface could be acquired.
Owing to the very low probability of a natural defect occurring in the descaled
strip, recourse was made to artificial defects, of similar dimensions to those
given above, stamped into the slab surface by means of a hammer and chisel.
The CCD camera, incorporating an infra-red blocking filter, together with a
high intensity sodium lamp for artificial illumination of the slab surface,
gathered image data which was stored on a computer disc. This disc, when read
into a laboratory computer with appropriate software successfully displayed the
artificial defects on a VDU screen. One CCD camera, of 2048 elements, was
considered sufficient for the broad face of the slab for the defect size of
interest. However, owing to the complexity of coupling such a camera to an image
analyser and recognising the fact that only a consecutive sequence of scans across
a slab could give defect information (a feature which is inherent in a video
camera) trials using a video camera were carried out.
The artificial defects were again employed to prove the video camera technique.
Infra-red blocking filters and artificial illumination of the slab surface by
means of high power tungsten halogen lamps were both necessary. These defects
were successfully extracted from the recorded images by means of an image analyser
system.
Some transverse corner cracks in a cold and moving sample were also successfully
detected. Finally, natural corner cracks were recorded and extracted. FR 107-9 842 721.GA/605
The unexpected difficulties associated with defect detection led to the link
between a detection system and the rectification equipment not being made.
However, the concept of using an inexpensive 'robotised' lorry loader for a
steelworks task, such as carrying a scarfing torch, has been proven.
Although the detection-rectification link was not made, the computer controlled
loader was put through a series of tests to prove its accuracy and its ability to
be 'decoupled' i.e. to move vertically or, more importantly, horizontally for
spot scarfing.
Mechanical modifications carried out to the standard loader were the replacement
of a single acting by a double acting hydraulic cylinder, the replacement of a
motor/gear drive unit on the slew motion by a hydraulic cylinder and crank
arrangement, and the fitment of positional and rate transducers.
Two techniques of control were investigated. A vectored rate control concept in
which the torch moved from the start to the end of a 'defect' at a constant rate
was compared with a 'dog-race' method by which the torch moves to a series of
'set-points' along the 'defect' path. The 'dog-race' technique was considered
superior despite the variable 'scarfing rate' since it did not suffer from
inaccurate trajectories inherent in the vectored rate control. These occurred at
the position to rate switchover at the start of a 'defect'. FR 107-9 842 7210.GA/605
CONTENTS Page
1 INTRODUCTION 1
1.1 Inspection Activities
1.2 Selective Scarfing By Robotics 2
1.3 Automatic Rectification Of Defects
1.4 Dye Penetrant Method , 2
2 SURFACE INSPECTION OF HOT SLABS . 3
2.1 Crack Detection In The Laboratory 3
2.2 Plant Installation 4
2.3 Camera Resolution On Hot Slabs
2.4 Line Scan Camera Operation
2.5 Data Analysis 5
2.5.1a Display
2.5.2 Data Processing
2.6 Results From Rolled Slabs
3 HOT INSPECTION OF CONTINUOUSLY CAST SLABS 6
3.1 Performance Of The Line Scan Camera
3.1.1 Laboratory Evaluations
3.1.2 Groove Roll Evaluation 7
3.1.3 Illumination Sources4 Processing Of The Line Scan Data 8
3.2 Plant Trials Using The Linescan Camera
3.2.1 Camera And The Artificial Illumination Arrangement2 Line Scan Camera Results 9
3.2.3 Method Employed To Record Real Defects 10
4 VIDEO CAMERA ASSESSMENT1
4.1 Strand Surface Conditions
4.2 Video Camera Installation 1
4.3oa Results
4.3.1 Imaging Techniques2 Contour And Colour Synthesizer Trials2
4.3.3 Comparison Of Video And Line Scan Cameras Using
Artificial Defects
5 OPTICAL FOURIER FILTERING AND THERMAL IMAGING CAMERAS 1
5.1 Optical Fourier Filtering
5.2 Thermal Imaging Camera3
6 IMAGE PROCESSING 1
6.1 Choice Of Equipment
6.2 Crack Detection Techniques
6.3 Image Processing Results4
6.3.1 Artificial Longitudinal Defects2 Transverse Corner Crack Detection - Laboratory Assessment 1
6.3.3 Image Processing Requirements
For A Permanent Plant Installation5
m FR 107-9 842 7210.GA/605
CONTENTS Page
7 HIGH PRESSURE WATER DESCALING 15
7.1 Installation
7.2 Slab Distortion6
7.2.1 Action To Prevent Slab Distortion
7.2.2 Descaling Effectiveness7
8 ROBOT DEVELOPMENT8
8.1 Objectives 1
8.2 The Requirements Of A Robotic Device For Selective Scarfing 1
8.3 Model Robot Arm9
8.3.1 Mechanics And Control2 Open Loop Control
8.3.3 Closed Loopl
8.4 Description Of Lorry Loader Arm 20
8.5 Concepts Of Loader Control
8.6 Mechanical Modifications To The Loader
8.7 Analogue Control Of The Robot Rams1
8.8 Performance Of The Computer Controlled Arm
8.8.1 Three Dimensional Position Control Repeatability 22 Ram Rate Control 2
8.8.3 Vectored Rate Control
8.8.4 Dog-Racel2
8.8.5e Versus Vectored Rate Control
8.9 Concept Of Robot Movement During Scarfing 23
9 CONCLUSIONS
10 REFERENCES 2
TABLES5
FIGURES 3
ADDENDUM 93

Univers
Ebooks
Livres audio
Presse
Podcasts
BD
Documents

Slab surface inspection and rectification

YouScribe

Le catalogue

Le service

Les conditions