Project Info

Machine learning to optimize additive manufacturing parameters

One fundamental theory underpins materials science: process dictates structure and structure dictates properties. However, predicting this process—structure—process or PSP relationship is elusive, because it requires knowledge of the orientation of every grain, the location and distribution of defects, and every processing condition that lead to this structure, down to the smallest fluctuations in temperature, pressure and composition. From an informatics perspective, this creates an extremely large input space. Weak and redundant variables and variables that cannot be measured convolute important, measurable values. However, machine learning provides a number of robust techniques to extract PSP relationships from these convoluted data streams. Approximately 6000 samples have been printed to characterize the build parameters for 3D printed metal samples. The tested samples connect the microstructure and mechanical properties to laser power, speed, spot size, powder size, shape and part orientation. These data serve as the basis for development of machine learning (ML) algorithms — including decision trees, scalable vector regression, and random forest networks — that focus on two-way modeling of process-property and process-structure relationships. Our results show how these parameters effect mechanical performance through microstructure, particularly keyhole
and lack-of-fusion porosity defects. This project will focus on the development of a data collection, processing, validation and distribution framework; on ML performance, accuracy and validation procedures.

More Information

Some reference papers are provided below:
https://arxiv.org/pdf/1710.10324.pdf?utm_source=Data-Driven+Materials+Science+Newsletter&utm_campaign=bd68f0f906-EMAIL_CAMPAIGN_2017_11_05&utm_medium=email&utm_term=0_60e6d7fee0-bd68f0f906-297887237
https://arxiv.org/pdf/1711.00118.pdf?utm_source=Data-Driven+Materials+Science+Newsletter&utm_campaign=bd68f0f906-EMAIL_CAMPAIGN_2017_11_05&utm_medium=email&utm_term=0_60e6d7fee0-bd68f0f906-297887237
https://pubs.acs.org/doi/pdf/10.1021/acs.chemmater.7b03500
https://pubs.acs.org/doi/pdf/10.1021/acs.jpclett.7b02333
https://arxiv.org/pdf/1710.03319.pdf?utm_source=Data-Driven+Materials+Science+Newsletter&utm_campaign=4611562463-EMAIL_CAMPAIGN_2017_11_05&utm_medium=email&utm_term=0_60e6d7fee0-4611562463-297887237

Grand Engineering Challenges: Engineer the tools of scientific discovery

Student Preparation

Qualifications

Students are expected to take CSCI 261, 262 before taking this project. It would be good if the student have already taken CSCI 303, 358, 404, 470, but this is not required.

Time Commitment

40 hours/month

Skills/Techniques Gained

1. The students will learn the skills to perform data processing and management.
2. The students will be involved my research team to perform research on machine learning and data mining.
3. The students will be involved into scientific paper writing for the results from this project.
4. The students will have chance to work together with my collaborators in industry.
5. The students will gain the fundamental knowledge on additive manufacturing, as well as how to use machine learning, as well as computational algorithms, to deal with problems in advanced material manufacture.
In a word, after the training in this project by successfully completing the assigned research tasks, the student is expected to be ready for pursing a graduate degree in the area of machine learning, data mining, or artificial intelligence, or a broader area of computer science.

Mentoring Plan

1. One orientation meeting is planned at the beginning of the project, in which the undergraduate students will be introduced to the research team. The project and research culture of the faculty’s research team will be introduced to the undergraduate students in the meeting.
2. Technical seminars within the research team are planned, once per week. In the every meeting, the undergraduate students will present a research paper relevant to the project and lead discussions on it with the faculty and the graduate students in the research team.
3. Professional development sessions within the research team are planned, once per week. In every meeting, the faculty or the graduate students in the research team will examine the progress of the project and the recent research results, exchange the ideas with the undergraduate students, and help them
develop research skills, including algorithm development, experimental design, scientific results evaluation, paper writing, and so on.
4. A poster session will be conducted at the end of the project in which the results of this project will be presented to the research teams of the faculty, the Computer Science Department, and the collaborators of the faculty.