Presentation on

Research Progress Seminar (RPS) Report

To Investigate and Analyze of Temperature

Profile in Additive Manufacturing
Presented by
Shaikh Tauseef
(Research Scholar )
PRN 191101354011
Department of Mechanical Engineering
School of Engineering and technology
Sandip University, Nashik

Guided by
Dr. Amol D. Lokhande
Associate Professor in Mechanical Engineering,
Sandip School of Engineering and Technology, Nashik
 Highlights of RPS

In this RPS
➢ The process of rapid prototyping has been studied.
➢ Various methods of Temperature Profile Prediction
have been discussed.
➢ Advantages, achievements and limitations of various
methods have been short listed .
➢ Problem is identified and research gap has been stated
clearly.
➢ Methodology for the work is proposed.
➢ Working model is also prosed.
 TABLE OF CONTENTS

1. INTRODUCTION

2. LITERATURE REVIEW

3. CHALLENGES

4. PROPOSED METHODOLOGY

5. BLOCK DIAGRAM

6. REFERENCES
 INTRODUCTION

✓Additive Manufacturing (AM) is a modern manufacturing approach

in which digital 3D design data is used to build parts by
sequentially depositing layers of materials.
✓ AM requires minimal or no tooling which reduces the cost
associated with conventional manufacturing processes.
✓Selective laser melting (SLM) is a widely used metal additive
manufacturing (AM) process, in which high-density laser is used to
fully melt and fuse metal powders to build parts in a layer-by-layer
manner.
✓The powder bed fusion (PBF) process for metal additive
manufacturing is gaining increasingly widespread recognition for its
ability to produce components with complex geometry.
 INTRODUCTION (cont.)

Why temperature monitoring and control are required in AM

Due to the large thermal gradient from repeatedly rapid heat and
solidification in AM during the process, defects such as porosity and cracks,
undesired residual stress , fatigue, and part distortion are frequently observed
in the produced parts. Therefore, temperature monitoring and control are
required in AM .
Some Researcher have used Infrared (IR) camera
Infrared (IR) camera can measure the temperature profile only on the
exposed surfaces i.e. outer surfaces.
Some have used Contact Methods
Contact methods such as the embedded thermocouple and non-contact
methods such as infrared pyrometer and infrared camera are used for the in-
situ temperature measurements.
I am proposing Machine Learning approach
Above all, the machine learning approach has been applied to a wide range
of areas, like strategies planning, optimization and calls for integration with
other fields
 LITERATURE REVIEW

S.NO Author Methods Pros Cons Achievements

1. Paul, A et A real-time iterative The models can be The effect of the The temperature
al[1] machine learning trained in an in situ size of the voxel profiles can be
Aug 2019 approach for method which is mesh on the predicted with a
temperature profile important for the temperature mean error value of
prediction versatility of the prediction results is less than 1%.
simulation process. not yet studied.

2. Roy, M. and Data-driven modeling computational cost, It is not applicable The accuracy
Wodo, O [2] of thermal history in power and storage in case of more obtained in this
June 2019 additive space required are complex method is greater
manufacturing very low. geometries. than 95% and it is
1000 times faster
when compared
with the physics
based models.
 LITERATURE REVIEW (CONT..)

S.NO Author Methods Pros Cons Achievements

3. Ren, K et al[3] Laser aided A wide variety of It cannot it achieves prediction

May 2019 additive scanning perform the accuracy of more
manufacturing strategies are thermal analysis than 95% when
(LAAM) to applicable in of multi-layered compared with the
generate training thermal field 3D deposition finite element model
data for a physics- prediction and process. physics-based
based machine also it performs machine learning
learning algorithm continuous algorithm
temperature field
prediction

4. Mirkoohi, E et Thermo- It involves fast The additional The thermal stress

al [4] mechanical and accurate models for and strains in metal
Nov 2019 modeling of prediction of the predicting the AM processes are
thermal stress stress state state of stress predicted in less than
and validating 3s
the models has
to be developed.
 LITERATURE REVIEW (CONT..)
S.NO Author Methods Pros Cons Achievements

5. Liang, S.Y et al a closed-form It obtains higher This method is not this model
[5] solution for the computational used for imposing operates at a
June 2019 temperature efficiency with the boundary frequency of about
prediction in the improved conditions on both 2.8 GHz and the
selective laser prediction the geometrical average
melting (SLM) accuracy. complex and computational
lateral boundary. time required is 12
s.

6. Ning, J et al [6] An efficient It requires low This method is not The average
April 2020 predictive model computational used for multi- computational
based on solid time. tracking and layer time is 217.93 s.
heat transfer for scanning.
temperature
profiles in
powder bed
metal additive
manufacturing
(PBMAM)
 LITERATURE REVIEW (CONT..)
S.NO Author Methods Pros Cons Achievements

7. Ahsan, F. and Two finite- It is a better method This method is The error in the
Ladani, L [7] element thermo to achieve not applicable in final
2019 fluid models temperature case of aluminum. concentration
prediction, bead obtained is -
shape as well as the 11.3427.
concentration of
alloying elements.

8. Mirkoohi, E a physics-based it increases the scan moving heat The error

et al [8] analytical model speed that tends to source approach obtained is less
Aug 2020 to predict increase the tensile for transient than 7.6 %
the in-process stress, it is reliable temperature field
thermal stress and valuable tool for measurement is
the prediction of the not applicable in
thermal stress in a case of high
faster manner. speeds as the
changes in melt
pool geometry
cannot be viewed
 CHALLENGES
• Voxel mesh size effect
The main issue encountered in, Real-time iterative machine learning
approach for temperature profile prediction [1] is that the extend of voxel
mesh size effect on the prediction results is not clearly described, so that the
prediction accuracy needs to be highly focused [1].

• Handling complex geometries

The prime challenge experienced in, Data-driven modeling of thermal
history in additive manufacturing technique [2], is that it is not applicable in
case of complex geometries, so that the input parameters have to be given in
a detailed manner in enabling the method to operate in more complex
geometries [2].

• Thermal prediction
The Laser aided additive manufacturing (LAAM) [3] method faces a
major challenge as it cannot perform the thermal prediction on the
multilayered 3D deposition materials [3].
 CHALLENGES (Cont.)
• Imposing boundary conditions
The closed-form solution for the temperature prediction
method cannot be used in case of imposing boundary
conditions on the geometrical complex boundary as well as
the lateral boundary which is considered as the significant
challenge encountered in this method [5].
• Thermo-mechanical modeling
The hectic challenge encountered in Thermo-mechanical
modeling of thermal stress [4] method is that it is not able to
predict the stress state within the manufactured part, so that
new models have to be developed with improved stress state
prediction abilities [4].
 Research Objectives

1) To study and analyse the existing models of prediction.

2) To design and develop an AI-based prediction model for
temperature profile prediction.
3) To propose a new AI-based algorithm to optimally tune the
hyper-parameters of the predictor.
4) To reveal the effectiveness of the developed model with a
comparative analysis based on the performance metrics.
Existing System
Data generation using
FEM

Transformation of 3-D
voxel data to a structured
data with neighboring
temperature and tool path

At each stage a new model is

created that uses the previous
values of the predicted stage

Ensemble model using

extremely randomized Iterative model
trees to predict next
i-time-steps

Block diagram of Existing system

 PROPOSED METHODOLOGY

❖ Ultimate goal is to proposed AI based Model:-

The ultimate goal of the research will be to design and develop a proposed
temperature profile prediction model based on the proposed optimized neural
network named Global herding algorithm based neural network (global herding
algorithm NN).

❖ It’s Novelty is: Global herding –based NN for prediction

The proposed global herding algorithm-based NN will be developed with the
integration of the herding characteristics of two algorithms namely, elephant
herding optimization [35] and rhino herding optimization [36] with the
NN model in such a way that the tunable parameters of NN will be optimally
decided so that the performance of the temperature profile prediction will be
enhanced. On the other hand, the dataset used in this proposal further boosts
the temperature profile prediction performance.

❖ Over Steps:- Thus, the overall steps in this research include:

1) Model fitting
2) Temperature prediction
 PROPOSED METHODOLOGY (CONT..)

❖ Iterative AI-based model is proposed

In this research, an iterative AI-based model is proposed for
predicting the data points of the voxels, which in turn predicts
the temperature profile of the data points.
❖ Develops the initial model
In order to predict the data points of the voxels, a ground truth table
is utilized, which develops the initial model, which is an elementary
model in prediction.
❖ Develop model for predicting the temperature profile
The next model for predicting the temperature profile of the voxels in
successive time-steps is evolved by integrating the data points
predicted by the initial model with the ground truth table.
 PROPOSED METHODOLOGY (CONT..)

❖ Furthermore, the temperature profiles in the upcoming time-steps are

obtained from the predicted temperature profiles and ground truth data
of the previous stages.
❖ The predicted data points at each iterative step will not be similar to
that of the data-points predicted for each voxel, due to the fact that not
the new voxels are predicted but also the temperature of the voxels in
the training set are predicted.
❖ Implementation Tool: PYTHON
❖ Performance metrics used for the analysis
❖ Note Overall time, and Mean Average Error (MAE)
❖ Models for Analysis
The comparative methods employed for the analysis of the proposed
method will be Surrogate Model (SM) [2], RNN–DNN thermal
analysis model [3] and closed-form solution model [5].
 BLOCK DIAGRAM

Fit proposed Predict Alternating fit Fit proposed

Fit proposed Predict and predict Predict final
NN model next k NN model on
NN model on next k …… stage k time-steps
on A time- time-steps t-l time-steps
first A+l time- time-steps
steps
steps

• Proposed Block diagram

Figure 1. Block diagram of NN- based temperature profile prediction

 REFERENCES

❑[1] Paul, A., Mozaffar, M., Yang, Z., Liao, W.K., Choudhary, A., Cao, J. and
Agrawal, A.," A real-time iterative machine learning approach for temperature
profile prediction in additive manufacturing processes", In proceedings of
2019 IEEE International Conference on Data Science and Advanced Analytics
(DSAA) ,pp. 541-550, October,2019.
❑[2] Roy, M. and Wodo, O.," Data-driven modeling of thermal history in
additive manufacturing", Additive Manufacturing, vol. 32, pp.101017, 2020.
❑[3] Ren, K., Chew, Y., Zhang, Y.F., Fuh, J.Y.H. and Bi, G.J.," Thermal field
prediction for laser scanning paths in laser aided additive manufacturing by
physics-based machine learning", Computer Methods in Applied Mechanics
and Engineering, vol.362, pp.112734, 2020.
❑[4] Mirkoohi, E., Sievers, D.E., Garmestani, H. and Liang, S.Y.," Thermo-
mechanical modeling of thermal stress in metal additive manufacturing
considering elastoplastic hardening", CIRP Journal of Manufacturing Science
and Technology, vol. 28, pp.52-67, 2020.
 REFERENCE (CONT..)
❑[5] Liang, S.Y., Ning, J.Q. and Mirkoohi, E.," A Closed-Form Solution for
Temperature Profiles in Selective Laser Melting of Metal Additive
Manufacturing", In Materials Science Forum ,Vol. 982, pp. 98-105, 2020.
❑[6] Ning, J., Wang, W., Ning, X., Sievers, D.E., Garmestani, H. and Liang,
S.Y.," Analytical thermal modeling of powder bed metal additive
manufacturing considering powder size variation and packing", Materials,
vol. 13 no.8, pp.1988, 2020.
❑[7] Ahsan, F. and Ladani, L.,"Temperature profile, bead geometry, and
elemental evaporation in laser powder bed fusion additive manufacturing
process",Jom, vol.72 no.1, pp.429-439, 2020.
❑[8] Mirkoohi, E., Dobbs, J.R. and Liang, S.Y.,"Analytical mechanics
modeling of in-process thermal stress distribution in metal additive
manufacturing", Journal of Manufacturing Processes, vol. 58, pp.41-54,
2020.
 REFERENCE (CONT..)

❑[9] Watson, J.K. and Taminger, K.M.B.," A decision-support model for

selecting additive manufacturing versus subtractive manufacturing based
on energy consumption", Journal of Cleaner Production, vol. 176,
pp.1316-1322, 2018.
❑[10] Ding, Y., Warton, J. and Kovacevic, R.," Development of sensing and
control system for robotized laser-based direct metal addition system",
Additive Manufacturing, vol. 10, pp.24-35, 2016.
❑[11] Russell, S. and Norvig, P., " Artificial intelligence: a modern
approach", 2002.
❑[12] Coveney, P.V., Dougherty, E.R. and Highfield, R.R.," Big data need
big theory too", Philosophical Transactions of the Royal Society A:
Mathematical, Physical and Engineering Sciences, vol.374, no.2080,
p.20160153, 2016.
 REFERENCE (CONT..)
❑[13] Arthur, J.B., "Effects of human resource systems on manufacturing
performance and turnover", Academy of Management journal, vol.37, no.3,
pp.670-687, 1994.
❑[14] Peyre, P., Aubry, P., Fabbro, R., Neveu, R. and Longuet, A.," Analytical
and numerical modelling of the direct metal deposition laser process", Journal
of Physics D: Applied Physics, vol.41, no.2, pp.025403, 2008.
❑[15] Everton, S.K., Hirsch, M., Stravroulakis, P., Leach, R.K. and Clare,
A.T.,"Review of in-situ process monitoring and in-situ metrology for metal
additive manufacturing", Materials & Design, vol.95, pp.431-445, 2016.
❑[16] Fu, C.H. and Guo, Y.B., "Three-dimensional temperature gradient
mechanism in selective laser melting of Ti-6Al-4V", Journal of Manufacturing
Science and Engineering, vol.136, no.6, 2014.
 REFERENCE (CONT..)
• [17] Fu, C.H. and Guo, Y.B., "Three-dimensional temperature
gradient mechanism in selective laser melting of Ti-6Al-4V",
Journal of Manufacturing Science and Engineering, vol.136,
no.6, 2014.
• [18] Atzeni, E. and Salmi, A.," Economics of additive
manufacturing for end-usable metal parts",The International
Journal of Advanced Manufacturing Technology,vol.62, no.9-
12, pp.1147-1155, 2012.
• [19] Slotwinski, J.A., Garboczi, E.J. and Hebenstreit, K.M.,"
Porosity measurements and analysis for metal additive
manufacturing process control", Journal of research of the
National Institute of Standards and Technology, vol. 119,
pp.494, 2014.
• [20] Yi, H., Qi, L., Luo, J. and Li, N., " Hole-defects in soluble
core assisted aluminum droplet printing: Metallurgical
mechanisms and elimination methods", Applied Thermal
Engineering, vol.148, pp.1183-1193,2019.
 REFERENCE (CONT..)
❑[21] Wu, A.S., Brown, D.W., Kumar, M., Gallegos, G.F. and
King, W.E.," An experimental investigation into additive
manufacturing-induced residual stresses in 316L stainless steel",
Metallurgical and Materials Transactions A, vol. 45, no.13,
pp.6260-6270, 2014.
❑[22] Mercelis, P. and Kruth, J.P.," Residual stresses in selective
laser sintering and selective laser melting", Rapid prototyping
journal, vol.12, no.5, pp.254-265, 2006.
❑[23] Yavari, S.A., Wauthlé, R., van der Stok, J., Riemslag, A.C.,
Janssen, M., Mulier, M., Kruth, J.P., Schrooten, J., Weinans, H.
and Zadpoor, A.A.," Fatigue behavior of porous biomaterials
manufactured using selective laser melting", Materials Science
and Engineering: C,vol.33, no.8, pp.4849-4858, 2013.
❑[24] Van Hooreweder, B. and Kruth, J.P.," Advanced fatigue
analysis of metal lattice structures produced by Selective Laser
Melting", CIRP Annals,vol.66, no.1, pp.221-224, 2017.
 REFERENCE (CONT..)
• [25] Dunbar, A.J., Denlinger, E.R., Heigel, J., Michaleris, P., Guerrier,
P., Martukanitz, R. and Simpson, T.W.,"Development of experimental
method for in situ distortion and temperature measurements during
the laser powder bed fusion additive manufacturing process",
Additive Manufacturing, vol.12, pp.25-30, 2016.
• [26] Li, C., Liu, J.F., Fang, X.Y. and Guo, Y.B.," Efficient predictive
model of part distortion and residual stress in selective laser
melting", Additive Manufacturing, vol. 17, pp.157-168, 2017.
• [27] Ning, J. and Liang, S.Y., “Predictive modeling of machining
temperatures with force–temperature correlation using cutting
mechanics and constitutive relation", Materials,vol.12, no.2,
pp.284,2019.
• [28] Ning, J. and Liang, S.Y.," A comparative study of analytical
thermal models to predict the orthogonal cutting temperature of
AISI 1045 steel",The International Journal of Advanced Manufacturing
Technology, vol.102, no.9, pp.3109-3119, 2019.
 REFERENCE (CONT..)
❑[29] Liverani, E., Toschi, S., Ceschini, L. and Fortunato, A., "Effect of
selective laser melting (SLM) process parameters on microstructure
and mechanical properties of 316L austenitic stainless steel",Journal
of Materials Processing Technology, vol.249, pp.255-263,2017.
❑[30] Kruth, J.P., Froyen, L., Van Vaerenbergh, J., Mercelis, P.,
Rombouts, M. and Lauwers, B.," Selective laser melting of iron-based
powder",Journal of materials processing technology,vol.149 no.1-3,
pp.616-622, 2004.
❑[31] Alsalla, H.H., Smith, C. and Hao, L.," Effect of build orientation
on the surface quality, microstructure and mechanical properties of
selective laser melting 316L stainless steel", Rapid prototyping
journal,vol.24, no.1,pp 9-17, 2018.
❑[32] Mirkoohi, E., Sievers, D.E., Garmestani, H. and Liang, S.Y.,"
Thermo-mechanical modeling of thermal stress in metal additive
manufacturing considering elastoplastic hardening", CIRP Journal of
Manufacturing Science and Technology,vol.28, pp.52-67, 2020.
 REFERENCE (CONT..)
❑[33] Mirkoohi, E., Dobbs, J.R. and Liang, S.Y.,"Analytical
modeling of residual stress in direct metal deposition
considering scan strategy", The International Journal of
Advanced Manufacturing Technology, vol.106, no.9, pp.4105-
4121, 2020.
❑[34] Mirkoohi, E., Tran, H.C., Lo, Y.L., Chang, Y.C., Lin, H.Y.
and Liang, S.Y.," Analytical mechanics modeling of residual
stress in laser powder bed considering flow hardening and
softening", The International Journal of Advanced
Manufacturing Technology, vol.107 no.9, pp.4159-4172, 2020.
❑[35] Mahdavi, M., Mirkoohi, E., Hoar, E., Liang, S. and
Garmestani, H.," Prediction of the deformation behavior of a
selective laser-melted Ti-6Al-4V alloy as a function of process
parameters", The International Journal of Advanced
Manufacturing Technology, vol.107 no.9, pp.4069-4076, 2020.
 REFERENCE (CONT..)

❑[36] Li, J., Lei, H., Alavi, A.H. and Wang, G.G.," Elephant herding
optimization: variants, hybrids, and applications", Mathematics,vol. 8
no.9, pp.1415,2020.
❑[37] Wang, G.G., Gao, X.Z., Zenger, K. and Coelho, L.D.S.," A novel
metaheuristic algorithm inspired by rhino herd behavior", In Proceedings
of the 9th EUROSIM congress on modelling and simulation, EUROSIM
2016, the 57th SIMS conference on simulation and modelling SIMS 2016
vol. 142, pp. 1026-1033, Linköping University Electronic Press,
December, 2018.
• [38] Dr. Amol D. Lokhande, Amit V. Patil et al, Design and Development
of Clamping Fixture for Drilling of Boiler Tube Plate , International
Journal of Engineering Research and General Science Volume 4, Issue 2,
March-April, 2016 ISSN 2091-2730