Discover the world's research
- 20+ million members
- 135+ million publications
- 700k+ research projects
Join for free
Welding Certification for Autonomous Robotized
Welding
Tomas Maske
UiT – The Arctic University of Norway
Faculty of Engineering Science and Technology
Narvik, Norway
tomas.maske@gmail.com
Gabor Sziebig
UiT – The Arctic University of Norway
Faculty of Engineering Science and Technology
Narvik, Norway
gabor.sziebig@uit.no
Abstract— The increased production of more and more
complex products challenges the accuracy of manual welding,
and increase the time it takes to program automatic welding
systems. The main objective for the article is to explore if current
regulations and standards are able to accommodate the shift
from automatic welding to autonomous welding systems. To do
this, the most current and applicable standards have been
analyzed. The findings are that most of the current standards
have room to accommodate autonomous systems, given that the
correct safety precautions are taken.
Keywords— automatic, autonomous, certificate, GTAW,
robotic, standards, welding
I.
I
NTRODUCTION
Since the first spot welding robot where installed [1] at
general motors in 1962, and later the first dedicated arc
welding robot by OTC Japan in 1979 the field has evolved fast.
Historically, robotic arc welding was regarded as a complex
shift for most production companies. It has been looked upon
as a change that required a high volume of repetitive welds to
justify the investment. Moreover, that it required a highly
skilled operator and programmer to fine tune and monitor the
welding process. Robotic welding has long been profitable for
lager manufacturers, but has been more challenging for
medium sized and job-shop companies to justify. This is about
to change, a new report from Market Research Reports
suggests an annual growth of 6.09 % during 2014 – 2019 in the
global industrial welding robot marked [2]. They also state that
one of the main challenges is the awareness of robot welding at
a regional level or at the end user. What drives the marked is
the increased usage of industrial robots, primary in the
automotive industry. Whit the introduction the AWS
Certification Program for Robotic Arc Welding - Operators and
Technicians [3] (American welding society), it is hoped that
the threshold for investing in robotic welding will be lowered.
This also paves the way to shift from automatic robot welding
to autonomous robot welding. The shift to more autonomous
welding processes is a change that will move the responsibility
for the weld quality from a person and over to a computer
system. This poses challenges in how to control the quality and
who is responsible if anything goes wrong. Current standards
will be analyzed along with trends in this shift, and usability of
these will be questioned. One other question that needs answer:
who is responsible if the robot makes a not satisfactory weld?
Moreover, if that not satisfactory weld cusses an accident. A
resent tragic incident from Germany, where a robot
accidentally killed a worker by pressing him up against a metal
plate and crushing his chest [4] must act as a warning to the
force available in industrial robots. In addition, it must remind
us that one of the key factors in automation of manual labor is
the safety for the workers.
II. L
IMITATIONS
This paper only discuss the certificates regarding three
different types of electric arc welding, also referred as Shielded
Metal Arc Welding / SMAW, Gas Metal Arc Welding /
GMAW or Gas Tungsten Arc Welding / GTAW. Some of the
findings may be transferable to other welding types, but that is
not taken into consideration.
III. C
URRENT
C
ERTIFICATES
A. Manual, mechanical and automated welding
Manual welding, this is the basic form of welding. The
operator holds the torch. This allows the welder to be close to
the weld, and is able to control the speed, heat and feed rate.
Mechanized welding, the welding torch is mounted on a trolley
or other device. Thus, removing the welding operator from
direct contact whit the weld. This process is still under the
direct control of the operator, so he / she must supervise and
adjust the speed of the torch, alternatively the oscillation, heat
and feed rate. This is similar to manual welding. However, the
operator is removed from direct contact whit the weld. This
allows for the usage of higher speed and greater heat. Due to
the less strenuous conditions for the operator, this also allows
for longer welds. Booth in time and distance. Automatic
welding, when an operator programs an automatic trolley, or a
robot, to follow a path and a given set of parameters. Whit inn
these parameters the apparatus is allowed to adjust its own
settings. The tolerance for error in the pieces that are assembled
are relatively low, and the torch can be snagged on the
imperfections.
B. Welding certificates
According to gowelding.org [5] welders are certified in
structural / plate and pipe welding. Inside those categories,
there exist a coding system to identify what kind of position /
orientation the welder is qualified to perform. For structural
welding the numbers 1 to 4 and the letters F and G is used: 1
stands for the flat position, 2 stands for the horizontal position,
3 stands for the vertical position 4 stands for the overhead
position, F stands for a filler weld joint and G stands for a
groove weld joint. This means that a 4F is a vertical weld done
using a filler joint. When it comes to structural certifications in
particular, groove welds will also qualify the welder for fillet
welds. However, fillet welds do not qualify the welder for
groove welds. Most codes allow a welder to take a combination
of the 3 and 4G positions, which typically qualifies the welder
for all position structural welding plus pipe that is a minimum
of 24 inches in diameter. Inn pipe welding the numbers 1, 2, 5,
6 and letters R, F and G is used. 1 is for a pipe in the horizontal
position that is rolled, 2 is for a pipe in the fixed vertical
position, 5 is for a pipe in the fixed horizontal position, 6 is for
a pipe in a 45 degree fixed position. R is for the restricted
position, F is for a fillet weld and G is for a grove weld. Here a
combination of 2 and 5G is used to prove that the welder is
qualified to weld in all pipe positions. The R limits the clearing
around the weld spot, and forces the welder to work within a
narrow space; it also forces the welder to use both hands. In
addition, ISO – 9606 – Qualification testing of welders —
Fusion welding [6], certifies welders to work whit different
materials. It consists of five parts. Part 1: Steels, part 2:
Aluminum and aluminum alloys, part 3: Copper and copper
alloys, part 4: Nickel and nickel alloys and part 5: Titanium
and titanium alloys, zirconium and zirconium alloys. There
exists other standards. Some apply in one country or region,
such as Germany or the EU. However, the most used, and
universal accepted is the ISO and ASME (American society of
mechanical engineers) standards. According to TWI (the
welding institute) [7] the certificates obtained under AWS
(American welder society) and ISO criteria are usable in most
countries. In addition, there exist special standard regarding
welding on high-pressure components, or equipment for use in
nuclear reactors. These certificates are the same for manual,
mechanized or automatic welding. As mentioned, the AWS has
developed a certificate that apply for robot arc welding. For
now this is only a certificate that is available in northern
America. The curriculum and tests are being revised to better
harmonize whit the ISO standards. The purpose is to ensure
that the beholder is qualified to operate a robot. The main
reason for this is the higher complexity of a new welding robot
compare to a new automatic welding machine. This ensures
that the operator understands the welding process and the
complexity of the kinematics of a robot. Modern flexible
welding systems for job-shop setup can be made up of a seven-
axes robot, equipped whit a two-axes welding table.
C. Current Situation for Autonomous Robotized Welding
AWS has developed a certification program specified
aimed at robotic arc welding, this is as mentioned called
CRAW. They future categorized three levels of users: level 1
and 2 operator and technician. The specifications are lied out in
AWS QC19:2002, second print April 2009 [8]. The basis of
knowledge to obtain this certificate is made up by several
different AWS standards and other sources:
• AWS A3.0 Standard Welding Terms and
Definitions
• AWS B1.10 Guide for Non-destructive Inspection
of Welds
• AWS B1.11 Guide for Visual Welding Inspection
• AWS B5.1 Qualification Standard for AWS
Welding Inspectors
• AWS QC1 Standard for AWS Certification of
Welding Inspectors
• AWS WI, Welding Inspection
• AWS CM-00 Certification Manual for Welding
Inspectors
• AWS B2.1 Specification for Welding Procedure
and Performance Qualification
• AWS D8.8 Specification for Automotive and
Light Truck Weld Quality: Arc Welding
• AWS D16.2 Standard for Components of Robotic
and Automatic Welding
• AWS D16.3 Risk Assessment Guide for Robotic
Arc Welding
• AWS D16.4 Specification for the Qualification of
Robotic Arc Welding Personnel
• ANSI Z49.1 Safety in Welding, Cutting and
Allied Processes
• NEMA EW-1 Electric Arc Welding Power
Sources
• AWS Arc Welding with Robots, Do's and Don'ts
• Automating the Welding Process, Jim Berge,
Industrial Press
• AWS Welding Handbook, Volume 1, 9th Edition
• AWS Welding Handbook Volume 2, 8th Edition
• Robot Programming Manual (published by robot
manufacturer)
• AWS 058 Arc Welding Automation, Howard Cary
• AWS A2.4 Standard Symbols for Welding
Brazing, and Non-destructive Examination
• Jefferson's Welding Encyclopedia 8th Edition
• RIA 15.06 American National Standard of
Industrial Robots and Robot Systems – Safety
Systems
These standards only take into consideration the technical
side of autonomous robotized welding, it is assumed that
anybody who apply for this certificate is familiar to ANSI
Z49.1, Safety in Welding and Cutting, and Allied Processes
and RIA 15.06, American National Standard of Industrial
Robots and Robot Systems - Safety Systems.
IV. CRAW
AS
B
ASIS FOR
A
UTONOMOUS
W
ELDING
A. Personnel
AWS D16.4 Specification for the Qualification of Robotic
Arc Welding Personnel [9] describes the qualifications
required for the three different levels of CRAW certificates,
level 1 and level 2 operator and technician. When discussing
certified robotic welding there needs to be a department head
that is ultimately responsible for the weld quality of that plant
or department. In that regards it is wise to use the qualifications
of a CRAW –T (Certified Robotic Arc Welding Technician) as
a guideline. AWS D16.4 outlines the following (p.4):
1) Skills and Ability Requirements
• Have the ability to make changes to the weld data, torch
angles, electrode stick out, starting techniques, and other
welding variables. Have an extensive welding
background and a thorough understanding of the robotic
interfacing system.
• Demonstrate a thorough understanding of all aspects of
the robotic work cell. Demonstrate programming, robotic
arc welding, seam tracking, fixturing, and any other
welding or robotic related functions. Have the capability
to enter the work cell and make changes to the weld
program, main program, torch clean program, or any
other related programs. Capable of fixture changes to
improve part fit up and part locating.
• Be capable of performing file management tasks, such as
saving, copying, and deleting program files.
• Demonstrate expertise in the welding operations
including all of the arc welding robots, automated
welding equipment, and all manual welding operations.
• Be responsible for the initial weld inspection and be
familiar with the tools that measure the weldment
quality.
• Have the ability to perform weld cross sectioning by
cutting, polishing, and etching appropriate samples when
necessary.
• Keep accurate and up to date records, including issuing
revised weld procedures as needed.
2) Experience and Education Requirements
• Meet all of the experience and education requirements
from previous levels.
• Have a minimum of 3000 hours or 3 years arc welding
experience.
• Have a two year Associates Degree in
Welding/Robotics/Electrical or equivalent combination
of appropriate education and experience.
• Hold current CWI certification (Certified Welding
Inspector).
3) Training Recommendations
• Obtain training in the proper operation of cross
sectioning tools and related hardware such as plasma
cutting and band saws.
• Obtain instruction in the applicable destructive testing
methods used, such as macro etch or bend testing.
• Receive instruction in the operation of quality measuring
tools, including applicable computer software for
measuring weld cross sections.
• Complete programming courses offered by original
equipment manufacturers or equivalent robotic
programming courses.
• Become familiar with personal computers and relevant
software.
B. Terms and definitions
AWS A3.0 Standard Welding Terms and Definitions [10],
AWS A2.4 Standard Symbols for Welding Brazing, and Non-
destructive Examination [11]. These standards ensures that
everybody that is welding and / or programing welding
machinery are talking the same language. This is to minimize
errors and misunderstanding due to miscommunication. It is
important that this form the basis for all machine human
communication. They also outline the symbols that should be
used in the design process. Using the same symbols means that
humans and / or machine will use correct welding methods.
C. Inspection
AWS B1.10 Guide for Non-destructive Inspection of Welds
[12] and AWS B1.11 Guide for Visual Welding Inspection
[13]. B1.10 outlines several methods for inspecting a weld
without causing any structural damage. The methods covered
in the standard is visual, liquid penetrant, magnetic particle,
radiographic, ultrasonic, electromagnetic (Eddy Current), leak.
In addition, the methods for visual inspection are described in
B 1.11. It states that currently x-ray is the fastest method for
inspecting welds, but it has a problem detecting deformations
going parallel whit the weld plane. Ultrasonic inspection is
regarded at the best method, and using a multi beam ultrasonic
device will most likely detect all deformations in any direction.
The information in these standards needs to be in cooperated in
an automated system for non-destructive testing of welds.
Wenfei Chen, Zuohua Miao and Delie Ming [14] has created
an x-ray based inspection tool that is able to inspect welding
line of steel tubes at "production line speeds". Whit the
implantation of high speed processing of machine vision and
new algorithms they successfully detected the same fault as a
manual inspection. In addition, the system had the capacity to
inspect every weld, and not only a selection. Due to the
unhealthy outcome from human exposure to x-rays, it is also a
good alternative to use ultrasound. Gordon Dobie, Walter
Galbraith, Charles MacLeod, Rahul Summan, Gareth Pierce
and Anthony Gachagan [15] have produced good results whit
ultrasonic technology. They were not able to reproduce the
speeds that x-ray are able to, that is mostly due to the usage of
Wi-Fi to upload the images. The maximum detection speed in
this device was 20 mm/s. Moreover, this product is developed
for an autonomous unit for inspection inside pipes. Mounting
this devise on a robot, whit a cabled data connection would
solve this problem. Both these solutions are interesting in the
scope of certified robotic welding.
AWS B5.1:2013 [16], AWS QC1 [17], AWS WI [18] and
AWS CM:00 [19] outlines the qualifications and demands on a
welding inspector. It is this authors belief the even the most
advanced systems are still designed by humans, and thus being
a victim of human error. One system cannot be allowed to
control itself. This means that highly skilled human inspectors
must do some sort of random control of the welds.
Furthermore, the information must me in cooperated in the
quality system for the inspection of welds.
D. Parameters
AWS B2.1 Specification for Welding Procedure and
Performance Qualification [20] outlines the parameters for
welding carbon steel to austenitic stainless steel in the
thickness range of 0.82 mm2 through 5.26 mm2 filler wire
using gas tungsten arc welding (TIG). It cites the base metals
and operating conditions necessary to make the weldment, the
filler metal specifications, and the allowable joint designs for
fillet welds and groove welds. AWS D8.8M Specification for
Automotive and Light Truck Weld Quality: Arc Welding [21]
further outline the requirements for making an approved weld
for the automotive industry.
Inn all arc welding there is a need for a power sours,
the specifications for these is found in NEMA EW-1 Electric
Arc Welding Power Sources [22]. Depending on the material,
the thickness and depth of the weld, the settings on the power
supply have a great impact on the result. In DC welding, we
differ between negative and positive electrode. When using a
positive electrode we usually get a deeper penetration than a
negative electrode. However, a positive electrode has a higher
melt of, and therefore a higher depositing rate. In AC welding,
we get the benefits for both these types of welding. In a more
advanced setup, it is possible to manipulate the ratio of
polarity, frequency and even the form of the curve (sine /
square). Giving the welder the possibility to manipulate the
width of the arc and the penetration. We also can differ
between the current in the negative and positive phase. For
instance, frequency and waveform manipulation can be used
for cleaning / removing the oxide coating when welding
aluminum.
Jefferson's Welding Encyclopedia 8th Edition [23], this
book contains information on different materials. The
information includes data on melting point, heat transfusion
and heat distortion. This information can be combined whit
data gathered from automatic metal classification. In
combination whit automated measuring, this would allow a
robotic weld system to calculate the maximum transferee rate
of filler material. Eranga Ukwatta and Jagath Samarabandu
[24] have achieved a 99 per cent identification rate using
Laser-induced Breakdown Spectroscopy and a high definition
video camera. This method utilizes an artificial neural network
and a set of metal samples to "train" the system to recognize
different metals. The system is cheaper and easier to run on
computer systems than the exiting spectral analyses. In
addition, it is less susceptible to pollutions and impurities.
In an autonomous welding system the software have to
device a welding procedure specification (WPS) or weld recipe
by itself. The document contains detailed information, included
but not limited to, the materials, position of welding, filler
material, shield gas, flow rate, number of passes, welding
current, pre-heat temperature. Kranendonk has developed a
software called RinasWeld [25] for just this purpose. By
importing a CAD-drawing into the software, it can identify the
weld path, create a WPS and use this information in an offline
program for a robot. The software can utilize multiple robots at
the same time, generating a 100 % collision free robot path.
To ensure that the WPS produces good welds, a Procedure
Qualification Record (PQR) have to be produced. This is a
proof test of the weld recipe (WPS) is able to make a weld that
has the strength required. This is a proof that the WPS is valid
and ready to be used in production.
Normally a Welder Performance Qualification Record
(WPQ) also have to be produced. This document states that the
welder is capable to follow the WPS and produce good welds.
This will not apply for an autonomous welding system. This is
based on the fact that robots will copy every movement exactly
like they did on the first run.
E. Components
AWS D16.2 Standard for Components of Robotic and
Automatic Welding [26] defines what is needed in order for a
system to qualify as a robotic arc welding installation. Robotic
arc welding systems consist of a manipulator, power source,
arc welding torch and accessories, electrode feed system, de-
reeling system, shielding gas delivery system, welding circuit,
shielding and communication control, and grounding system.
Note that the standard does not require a safety system.
However, RIA 15.06 American National Standard of Industrial
Robots and Robot Systems – Safety Systems [27] and AWS
D16.3 Risk Assessment Guide for Robotic Arc Welding [28]
gives a good introduction into what systems needs to be in
place to qualify for a safe workplace. RIA 15.06 also form the
basis for ISO 10218 Robots and robotic devices - Safety
requirements for industrial robots.
F. Safety
ANSI Z49.1 Safety in Welding, Cutting and Allied
Processes [29], RIA 15.06 and AWS D16.3 have good
guidelines for what must be in place and what should be in
place. The RIA 15.06 is considered a game changer for the
robotic industry, it introduces some new concepts. In addition,
a welding cell must be covered by a screen, or similar
contraption, to shield people from the UV-light produced by
the welding process.
1) Functional safety
The goal of implementing functional safety is to define, as
well as quantify, engineering solutions (safety measures,
techniques and procedures) that need to be implemented to
achieve an acceptable safety hazard level in compliance with
the safety standard. In other words: Supplied components and
their integration into the safety-related control system must
meet the required safety performance level and have the life
expectancy needed to meet the system's overall functional
safety.
2) Safety-related Soft Limits (SRSL)
Historically, robotic safety and safeguarding was all about
hardware-controlled limits to a robot's movements, combined
with access restrictions to the potential motion space. When
ordering a "new" robot with the proper Safety Rated Soft Limit
and/or manufacturer hard stop options the system can be
programmed to use a smaller portion of the robot's maximum
reach area. By doing so the restricted space can be reduced to
closer match the shape of the required work envelope. Thus,
less perimeter safeguarding can be used and the guarding will
enclose less floor space. Now that SRSL's are safety-rated and
accepted by national standards. This can be of great benefit
since it allows further optimization of floor space. The overall
floor space required by the robotic system is reduced by
integrating the proper safeguard devices into your safety
control system.
V. C
ONCLUSION
The process of going from automated to autonomous
welding may be both a technical and political challenge. It is
the authors opinion that one of the best way to start is to begin
in parts of the industry where welding is considered to be
especially hazardous. Meaning welding inside tanks and "hard
to reach" places. The software and hardware solutions have
evolved to the point that it is capable of taking decisions that
will result in a high quality weld. One more important aspect is
to keep humans in the loop in a supervision capability.
Moreover, given the proven track record of robots in the
industry and the safety guard in place. There is a good chance
that we will see the first certified autonomous robotic welding
systems in near future.
A
CKNOWLEDGMENT
The authors wish to thank the support and financing to the
Research Council of Norway (BIA 245691, CoRoWeld).
R
EFERENCES
[1] RobotWorx, "www.robots.com," RobotWorx, Unknown. [Online].
Available: https://www.robots.com/education/industrial-history.
[Accessed 20 Januray 2016].
[2] Market Research Reports Inc,
"http://www.marketresearchreports.com/," 23 September 2015. [Online].
Available: http://www.marketresearchreports.com/technavio/global-
industrial-welding-robots-market-2015-2019. [Accessed 20 January
2016].
[3] The American Welding Society, "www.aws.org," Unknown. [Online].
Available: https://www.aws.org/certification/detail/certified-robotic-arc-
welding. [Accessed 20 January 2016].
[4] B. Brumfield, "http://edition.cnn.com/," CNN, 02 July 2015. [Online].
Available: http://edition.cnn.com/2015/07/02/europe/germany-
volkswagen-robot-kills-worker/index.html. [Accessed 20 Januray 2016].
[5] Go welding, "www.gowelding.org," Unknown. [Online]. Available:
http://www.gowelding.org/welding-certification/. [Accessed 18 January
2016].
[6] The International Organization for Standardization, "ISO 9606-1:2012 -
Qualification testing of welders - Fusion welding," The International
Organization for Standardization, 2012.
[7] T. J. J. D McKeown, "http://www.twi-global.com/," 28 April 1999.
[Online]. Available: http://www.twi-global.com/technical-
knowledge/published-papers/recent-developments-in-welding-
inspection-certification-including-developments-in-ewf-and-iiw/.
[Accessed 19 January 2016].
[8] The AWS Certification Committee, Specification for AWS Certification
of Robotic Arc Welding Personnel AWS QC19:2002, Miami: American
Welding Society, 2009.
[9] The American Welding Society Inc, AWS D16.4:2005 Specification for
the Qualification of Robotic Arc Welding Personnel, Miami: American
Welding Society, 2005.
[10] The American Welding Society Inc, ANSI/AWS A3.0-94 Standard
Welding Terms and Definitions, Miami: American Welding Society,
1994.
[11] The American Welding Society Inc, AWS A2.4:2012 Standard Symbols
for Welding Brazing, and Non-destructive Examination, Miami:
American Welding Society, 2012.
[12] The American Welding Society Inc, AWS B1.10:2009 Guide for the
Nondestructive Examination of Welds, Miami: American Welding
Society, 2009.
[13] The American Welding Society Inc, AWS B1.11:2015 Guide for the
Visual Examination of Welds, Miami: American Welding Society, 2015.
[14] Z. M. a. D. M. Wenfei Chen, "Automated Inspection using X-ray
Imaging," Trust, Security and Privacy in Computing and
Communications (TrustCom), 2011 IEEE 10th International Conference
on, pp. 1769 - 1772, 16 - 18 November 2011.
[15] W. G. C. M. R. S. G. P. A. G. Gordon Dobie, "Automatic Ultrasonic
Robotic Array," Ultrasonics Symposium (IUS), 2013 IEEE
International, pp. 1861 - 1864, 21 - 25 July 2013.
[16] The American Welding Society Inc, ASW B5.1:2013 Specification for
th Qualification ofWelding Insperctors, Miami: American Welding
Society, 2013.
[17] The American Welding Society Inc, AWS QC1:2007 Standard for AWS
Certification of Welding Inspectors, Miami: American Welding Society,
2007.
[18] The American Welding Society Inc, AWS WI:2015 Welding Inspector
Handbook, Miami: American Welding Society, 2015.
[19] The American Welding Society Inc, AWS CM:00 Certification Manual
for Welding Inspectors, Miami: American Welding Society, 2000.
[20] The American Welding Society Inc, AWS B2.1:2015 Specification for
Welding Procedure and Performance Qualification, Miami: American
Welding Society, 2015.
[21] The American Welding Society Inc, AWS D8.8M:2014 Specification
for Automotive and Light Truck Weld Quality: Arc Welding, Miami:
American Welding Society, 2014.
[22] National Electrical Manufacturers Association, NEMA EW 1-2004
Electric Arc-Welding Power Sources, Arlington: NEMA, 2004.
[23] R. L. O'Brien, Jefferson's Welding Encyclopedia, Miami: American
Welding Society, 1997.
[24] J. S. Eranga Ukwatta, "Vision Based Metal Spectral Analysis using
Multi-label Classification," Canadian Conference on Computer and
Robot Vision, 2009. CRV '09. pp. 132 - 139, 25-27 May 2009.
[25] Kranendonk, "http://www.kranendonk.com/," Kranendonk Production
Systems BV, Unknown. [Online]. Available:
http://www.kranendonk.com/software/rinasweld. [Accessed 27 January
2016].
[26] The American Welding Society Inc, AWS D16.2:2007 Standard for
Components of Robotic and Automatic Welding, Miami: American
Welding Society, 2007.
[27] Robotic Industries Association, ANSI/RIA R15.06:2012 Safety
Requirements for Industrial Robots and Robot Systems, Ann Arbor:
Robotic Industries Association, 2012.
[28] The American Welding Society Inc, AWS D16.3:2009 Risk Assessment
Guide for Robotic Arc Welding, Miami: American Welding Society,
2009.
[29] American National Standards Institute, ANSI Z49.1:2012 Safety in
Welding, Cutting and Allied Processes, Miami: American Welding
Society, 2005.
ResearchGate has not been able to resolve any citations for this publication.
-
![Eranga Ukwatta]()
- Jagath Samarabandu
Industrial equipments that employ element identification tend to be expensive as they utilize built-in spectroscopes and computers for post processing. In this paper we present an in situ fully automatic method for detecting constituent elements in a sample specimen using computer vision and machine learning techniques on Laser Induced Breakdown Spectroscopy (LIBS) spectra. This enables the development of a compact and portable spectrometer on a high resolution video camera. In the traditional classification problem, classes are mutually exclusive by definition. However, in spectral analysis a spectrum could contain emissions from multiple elements such that the disjointness of the labels is no longer valid. We cast the metal detection problem as a multi-label classification and enable detection of elemental composition of the specimen. Here, we apply both Support Vector Machine (SVM) and Artificial Neural Networks (ANN) to multiple metal classification and compare the performance with a simple template matching technique. Both machine learning approaches yield correct identification of metals to an accuracy of 99%. Our method is useful in instances where accurate elemental analysis is not required but rather a qualitative analysis. Experiments on the simulation data show that our method is suitable for LIBS metal detection.
- Wenfei Chen
- Zuohua Miao
- Delie Ming
With increasing concern on environmental contamination due to pipeline leak, the electronics industry is coming under increasing pressure to develop and apply automated inspection techniques for the inspection of welding line of steel tubes and structural casting. Automatic X-ray inspection systems are taking the high cost out of production inspection for casting manufacturers who previously relied on manual inspection methods while simultaneously wiping out the drudgery and potential for human error common to manual inspection methods in processing and manufacturing applications. Based on the analysis of basics of X-ray Imaging Principle, the interactive process of automatic X-ray inspection was discussed and a new defect inspection method using top-hat operator was put forward. Lastly, this method is applied for many samples of X-ray images, and proved to be effective.
- R L O'brien
R. L. O'Brien, Jefferson's Welding Encyclopedia, Miami: American Welding Society, 1997.
www.robots.com RobotWorx, Unknown. [Online] Available: https://www.robots.com
- Robotworx
RobotWorx, "www.robots.com," RobotWorx, Unknown. [Online]. Available: https://www.robots.com/education/industrial-history.
Automatic Ultrasonic Robotic Array
- W G C M R S G P A G Gordon Dobie
W. G. C. M. R. S. G. P. A. G. Gordon Dobie, "Automatic Ultrasonic Robotic Array," Ultrasonics Symposium (IUS), 2013 IEEE International, pp. 1861 -1864, 21 -25 July 2013.
06:2012 Safety Requirements for Industrial Robots and Robot Systems
- Ansi Robotic Industries Association
Robotic Industries Association, ANSI/RIA R15.06:2012 Safety Requirements for Industrial Robots and Robot Systems, Ann Arbor: Robotic Industries Association, 2012.
Unknown Available: https://www.aws.org/certification/detail/certified-robotic-arcwelding
The American Welding Society, "www.aws.org," Unknown. [Online]. Available: https://www.aws.org/certification/detail/certified-robotic-arcwelding. [Accessed 20 January 2016].
Available: http://www.twi-global.com/technicalknowledge/published-papers/recent-developments-in-weldinginspection-certification-including-developments-in-ewf-and-iiw
- T J J Mckeown
T. J. J. D McKeown, "http://www.twi-global.com/," 28 April 1999. [Online]. Available: http://www.twi-global.com/technicalknowledge/published-papers/recent-developments-in-weldinginspection-certification-including-developments-in-ewf-and-iiw/. [Accessed 19 January 2016].
Kranendonk Production Systems BV, Unknown
- Kranendonk
Kranendonk, "http://www.kranendonk.com/," Kranendonk Production Systems BV, Unknown. [Online].
- B Brumfield
B. Brumfield, "http://edition.cnn.com/," CNN, 02 July 2015. [Online]. Available: http://edition.cnn.com/2015/07/02/europe/germanyvolkswagen-robot-kills-worker/index.html. [Accessed 20 Januray 2016].
Source: https://www.researchgate.net/publication/312262387_Welding_certification_for_autonomous_robotized_welding
Posted by: chasechasebanaage0270901.blogspot.com
0 Comments