Participant Support Costs REU Supplement for CPS: Synergy: An Integrated Simulation and Process Control Platform for Distributed Manufacturing Process Chains

Project: Research project

Project Details


Northwestern University Statement of Work – Ehmann

This proposal is in response to solicitation number NSF 16-549 for NSF's Cyber-Physical Systems program.

Professor Kornel Ehmann would serve as the project PI for all 3 years of the project. Professors Jian Cao, Wing K. Liu, and Gregory Wagner would serve as the project Co-PIs for all 3 years of the project.

The proposed research will establish the scientific and technological foundation for a futuristic manufacturing platform in a distributed network that seamlessly and efficiently integrates physical processes and numerical simulations in a fast predictive framework powered by next-generation cyber-physical infrastructure to control the desired part characteristics in terms of its attributes/properties such as microstructure, porosity, residual stresses, geometric and surface features, etc. The envisioned cyber-physical engineering platform will allow users to quickly determine the process parameters needed to yield the desired part attributes in a given process or process chain. The landscape of manufacturing is gradually becoming predominantly digital due to the emergence of Cyber Physical Systems. At the forefront of this innovation are processes in which geometrically and structurally complex parts based on digital models (CAD) can be produced. In spite of their flexibility, the functional properties and performance of the produced parts is difficult to predict. This research targets the formulation of a new processing platform in which part geometry, process parameters, numerical process simulation, and predicted mechanical behavior all interact via a common data structure for process control and certification. The part geometry (from CAD), analysis meshes and part properties will all be saved through an innovative volumetric representation providing a common data structure, optimized for performance on Graphical Processing Units (GPUs). Models capable of predicting material micro-structure during Direct Energy Deposition processes and residual stresses in sheet-metal components by Double Sided Incremental Forming (DSIF) will be formulated and used to control these two processes. The feasibility of these models and of the proposed platform to achieve the desired part properties will be demonstrated on two unique fully-instrumented open-architecture networked machines. Keywords: Cyber Physical Systems, Direct Energy Deposition (DED), Double Sided Incremental Forming (DSIF), Numerical Simulation, Volumetric Representation.

First, the proposed research represents the inception of a fundamentally new volumetric representation capable of efficiently storing geometric, numerical simulation, and part attribute information coupled with very efficient modeling techniques, and infused in a framework with native support for massively parallel processing. Second, this research includes a methodology for efficiently predicting part attributes using new reduced-order models developed from fine scale full-order simulations. Third, the new framework provides a means by which manufacturing processes parameters can be optimized digitally and analytically to produce high performance parts using virtual experimentation and simulation. Fourth, a multi-loop control structure with the requisite real-time sensing, networking and control algorithms demonstrates the ability of the platform to control part attributes in two representative processes (DED and DSIF). The proposed work will be accomplished collaboratively by a team of researchers specialized in parallel computing, multi-scale c
Effective start/end date12/1/1611/30/20


  • National Science Foundation (CMMI-1646592-001)

Fingerprint Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.