FACAM 2020 Workshop will be hosted at USC on February 10th and 11th.

We may have limited seats available. Please contact Dr. Huang (qiang.huang@usc.edu).

Software Demo -- Machine Learning for Additive Manufacturing (ML4AM)

The software demo of Smart 3D Printing System Calibrator for Shape Accuracy Control has been posted.  Thanks the effort of PhD students Nathan and Yuanxiang.

A New Machine Learning for Additive Manufacturing (ML4AM) Project on Shape Deviation Generator and Learner

A new machine learning for AM (ML4AM) project, "NSF CMMI-1901514: Shape Deviation Generator and Learner- An Engineering-Informed Convolution Modeling andLearning Framework for Additive Manufacturing Accuracy Control", has been funded by US National Science Foundation. (PI: Huang, $350K, 08/2019∼ 07/2022).  The grant abstract is given below.

Although additive manufacturing (AM), known as 3D Printing, holds great promise as a direct manufacturing technology, significant deviations from the desired part shape often occur in the printed parts.  As a result, shape distortion control is a critical issue for AM built products. With advances in computing and increased accessibility of AM product data, machine learning for AM has become a viable strategy for enhancing 3D printing performance. However, meaningful learning of engineering data requires effective integration of domain knowledge, making general-purpose machine learning methods difficult to apply. As a result, there is a critical need for an engineering-informed, data-analytical, machine learning framework for shape distortion control.  Such a tool is essential to improving AM quality and reducing cost and waste.

The project will establish an engineering-informed convolution modeling and learning methodological framework for AM distortion control. A Shape Deviation Generator and Learner for shape accuracy control will be researched by: (1) modeling 3D shape deviation generation by establishing a new convolution formulation for layer-by-layer fabrication processes, (2) transferring the 3D shape deviation model from a small set of training shapes to a wider variety of shapes by exploring and learning shape similarity under a cookie-cutter modeling framework, (3) transferring the shape deviation model between AM processes by exploring and learning process similarity through an effect equivalence framework, and (4) validating modeling and transfer learning methodologies in both polymer and metal AM processes. Methodologies and tools will be developed to mitigate both shape and process complexities towards the goal of building AM products with high geometric fidelity: from single shape to multiple shapes, and from single process to multiple AM processes.

 

Position Opening at HP Labs, Palo Alto: Deep Learning Scientist for 3D Printing

HP Labs’ 3D Lab (3DL) invents next generation 3D printing technology and novel materials, and creates their digital twins to enable digital manufacturing.  A key technical foundation that delivers precision manufacturing at voxel level such as HP’s Multi Jet Fusion technology is ubiquitous sensing for both machine and products, and process control based on continuous sensing data stream enriched by physics.  HP’s Multi Jet Fusion 3D printers have been in production worldwide; each production build generates terabytes of data from geometrical processing to in-situ thermal sensing. 3DL has created a highly promising research in applying deep learning to these datasets –  a single production build of terabytes,  different builds for a single machine, and different builds by different machines in different geographical location.  We are expanding this deep learning research project to accelerate our preliminary success.  We are inviting you to join our journey. 

We expect you an accomplished deep learning specialist, in particular,

1. PhD (preferred) in engineering and/or applied physics from reputable universities and research programs. Multiple years of independent research with demonstratable outcome.
2. Hands on experiences in deep learning; experiences with recurrent neural network (e.g., LSTM) is a plus. Experiences with generative adversarial networks and/or reinforcement learning a plus.
3. Hands on experiences in established platforms; Tensorflow experience is a plus.
4. Hands on experiences in applying deep learning to analyze sensing data in the form of images, videos.
5. Research experiences on applying deep learning to solve scientific problems; some hardware/physical science background is a plus. 
6. Willing to lead research; willing learn and teach.

More details can be found from
https://h30631.www3.hp.com/job/palo-alto/machine-learning-scientist-for-3d-printing-research/3544/13031902

  Contact:  Dr. Zeng (Email: jun.zeng@hp.com)

Summary of FACAM 2019 Workshop

Co-organized by Prof. N. Anwer (ENS Paris-Saclay), Prof. C. Mehdi-Souzani (ENS Paris-Sacaly), and Prof. Q. Huang (USC), the third FACAM 2019 workshop was held on June 17th and 18th at ENS Paris-Saclay (Cachan).  Other than organizers, the participants include Dr. AF Obaton (LNE of French Metrology Institute), Prof. J. Liu (University of Arizona), Prof. R. Jin (Virginia Tech), Prof. M. Grasso (Politecnico di Milano), and graduate students  Zuowei Zhu, Yahya Al Meslem, Kevin Pereirrai (ENS Paris-Saclay), Baltej Rupal (ENS Paris-Saclay/University of Alberta), Saint-Clair Toguem Tagne (LNE/ENS Paris-Saclay), Charles Cayron(LNE), Yuanxiang Wang, Nathan Decker, and Chris Henson (USC).

Prof. Anwer opened the workshop by welcoming everyone and gave an introduction of the Advanced and Smart Manufacturing research at the LURPA research lab and AM Initiatives at ENS Paris-Saclay. Dr. Obtaton’s presentation introduced metrology challenges in AM and opportunities for statistical analysis and machine learning of metrology data. Prof. Grasso presented novel solutions for in-situ monitoring of metal AM. Baltej Rupal discussed geometric Benchmark test artifacts design for GD&T quantification. Dr. Huang presented a new convolution framework for learning and predicting shape accuracy in AM. Zuowei Zhu and Kevin Ferreira presented a data-driven Skin Model Shapes  and deep learning for AM. Nathan Decker ended the first day by presenting dimensional accuracy prediction using machine learning of triangular mesh data.  

On the second day of the workshop Prof. Liu presented the spatio-temporal modeling of photothermal curing process for glass printing. Prof. Jin discussed  data fusion for quality modeling in AM processes and network. Yuanxiang Wang, Nathan Decker, and Chris Henson presented a software prototype demo of cloud-based machine learning for 3D printing distortion control. In the afternoon session, Prof. Huang led the discussion on the collaboration opportunities and new research problems arising from the presentations and follow-up questions.  Prof. C. Mehdi-Souzani, Prof. Liu, Prof. Grasso, and Prof. Jin made great remarks (Prof. Anwer left earlier due to a conference). A whitepaper is expected to produced from the discussions. Prof. C. Mehdi-Souzani gave a warm concluding remark.

The FACAM 2020 workshop will be in USC. Looking forward to another productive and fun AM workshop!

Contact: Dr. Q. Huang (qiang.huang@usc.edu)

Photos from FACAM 2019 Workshop at ENS Cachan (updated)

        Co-hosted by Prof. N. Anwer, Prof. C. Mehdi-Souzani, and Prof. Q. Huang, the FACAM 2019 workshop was held on June 17th and 18th at ENS Cachan. 

        Below are a few photos from the Workshop courtesy of Baltej Rupal, Christopher Henson, Zuowei Zhu, Kevin, and Nathan Decker.  Informative presentations, delicious food, and wonderful people made for a fruitful two days of learning.

 

 

 

Nathan won the RAMP 2019 Competition!

A great achievement to win the RAMP 2019 award (https://event.asme.org/MSEC/Program/RAMP-2019-Competition) as a first-year PhD student! Congratulations to Nathan!

Hooding Ceremony for Yuan on May 9, 2019

Yuan comes back to USC from Facebook for the hooding ceremony today. Nice to have her in the lab.

Nathan becomes one of the three finalists for the RAMP 2019 competition!

Being a first-year PhD student at USC, Nathan will present  his paper at ASME MSEC 2019 conference and compete for the RAMP 2019 award (https://event.asme.org/MSEC/Program/RAMP-2019-Competition)!