Machine Learning Applications in Advanced Manufacturing Processes

  • PDF / 169,673 Bytes
  • 2 Pages / 593.972 x 792 pts Page_size
  • 105 Downloads / 294 Views

DOWNLOAD

REPORT


https://doi.org/10.1007/s11837-020-04380-5 Ó 2020 This is a U.S. government work and not under copyright protection in the U.S.; foreign copyright protection may apply

MACHINE LEARNING APPLICATIONS IN ADVANCED MANUFACTURING PROCESSES

Machine Learning Applications in Advanced Manufacturing Processes DONNA POST GUILLEN

1,2

1.—Idaho National Laboratory, 955 MK Simpson Blvd., Idaho Falls, ID 83415, USA. 2.—e-mail: [email protected]

Machine learning holds tremendous promise for revolutionizing modern manufacturing, from conventional operations to new advanced manufacturing processes, such as additive manufacturing. This special topic focuses on reducing waste, energy usage. and carbon emissions, spurring innovation in materials development and production. Advances in digital manufacturing, process control, predictive maintenance, and automation can be realized by the integration of data analytics and validated models to ensure product quality, optimize operations, enhance productivity, and improve efficiency. Articles within this special topic present mathematically- and computationally-based applications to improve the efficiency of industrial manufacturing processes. The papers featured in this special topic include ‘‘A Prediction Model of Blast Furnace Slag Viscosity Based on the Principal Component Analysis and KNearest Neighbor Regression’’ by Jiang et al., with a predictive model for the viscosity of blast furnace (BF) slag based on machine learning, principal component analysis, and k-nearest neighbor regression. Iron-making in a BF is one of the most complicated industrial processes, and viscosity has a large influence on the separation of BF molten iron and slag, the erosion of the hearth, and BF production. The detection of viscosity of BF slag has a large hysteresis, which confounds the

Donna Guillen served as the Guest Editor for the TMS Energy Committee and coordinated the topic Machine Learning Applications in Advanced Manufacturing Processes in this issue. Edward Herderick and Judy Schneider of the Additive Manufacturing Committee and Srikanth Patala of the Computational Materials Science and Engineering Committee provided useful discussions in the development and co-sponsorship of this special topic. (Received September 1, 2020; accepted September 9, 2020)

understanding of the slag blast status in real time. As a result of the rapid advancement of sensors and detectors, steel companies have accumulated a large amount of production data during long-term operation. A ML methodology was developed to use such data to provide the BF slag viscosity information in real time. ‘‘Assessment of Cutting Forces and Temperature in Tapered Milling of Stone–Plastic Composite using Response Surface Methodology’’ by Zhu et al. investigates the milling performance of a new engineering material as a replacement for traditional materials used in furniture, packaging, and flooring, due to its durability, wear resistance, and environmental friendliness. The machinability of stone–plastic composite (SPC) materials du