Page 13 - Moving Forward 2020: Harold and Inge Marcus Department of Industrial and Manufacturing Engineering magazine

“This work allows us us to to build a a a a a a a a a a a a a a a a a bridge between biological patterns patterns like like in in in in in in human anatomy and man-made patterns patterns like like in in in in in in manufacturing
” —Hui Yang associate professor of of industrial engineering To create the the novel algorithm the the team analyzed spatial data in in in complex microscopic images produced by ultra-precision machining (UPM) UPM UPM a a a a a a a a a a a manufacturing
process that uses single-crystal diamond tools to to to refine metal workpieces at at at the atomic scale is widely used in fin in in modern industries such as semiconductors and aerospace to to produce highly precise cuts or or polishing The spatial data showed a a a a a a a a a a variety of surfaces over the UPM images ranging from flat to to rough to to severely rugged Good quality quality products products should have have a a a a a a a a a a a a similar surface surface and bad quality quality products products might have have different textures on the surface surface This operation captured and reiterated the the the behaviors of recurrence variations in the the the spatial spatial data from the the the images to represent characterize and quantify spatial spatial patterns in the the UPM images The surface characteristics were shown to be highly correlated with with the the the spatial recurrence patterns within the imaging data According to to to Chen in in the the past researchers had to to to physically measure a a a a a a a a a a a piece to to to get the the quality of surface finishes when manufacturing
Their work now allows surface roughness to to be approximated by using the images which ultimately leads to to cost savings and resource conservation In the the future this methodology can improve predictive models for the the the quality quality of of UPM surface finishes to enhance the the quality quality of of manufacturing
“The algorithm teaches you new things about the system system as a a a a a a a a a a a a a a a a a whole ” Kumara said “Take for example:
a a a a a a a a a a a a a a a a a a a signal impulse happens in one part of a a a a a a a a a a a a a a a a a a a system system at at a a a a a a a a a a a a a a a a a a a given time time and and space space Later it it has an an an observed repetition at at at at a a a a a a a a a a a a a different point in in in time time and and space space If that pattern is found then you can use it to predict such behaviors in in in the the future ” According to to Yang the algorithm has broad implications for medical applications such as as monitoring organ organ function function analyzing cancer images and detecting organ organ dysfunction over time “You can use this algorithm on complex-structured data that is is is measurable or or or observable and is is is represented in 2D 3D or or high-dimensional images ” Yang said The National Science Foundation the the Allen Allen E E and and and Allen Allen M M M Pearce Endowment and and and and the the Harold and and and and Inge Marcus Department of Industrial and and and Manufacturing Engineering supported this work IME NEWSLETTER • VOLUME 5 2020
13
Research