| NSF Org: |
CBET Division of Chemical, Bioengineering, Environmental, and Transport Systems |
| Recipient: |
|
| Initial Amendment Date: | July 31, 2015 |
| Latest Amendment Date: | May 3, 2017 |
| Award Number: | 1547110 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Triantafillos Mountziaris
CBET Division of Chemical, Bioengineering, Environmental, and Transport Systems ENG Directorate for Engineering |
| Start Date: | January 1, 2016 |
| End Date: | December 31, 2018 (Estimated) |
| Total Intended Award Amount: | $235,869.00 |
| Total Awarded Amount to Date: | $259,869.00 |
| Funds Obligated to Date: |
FY 2016 = $12,000.00 FY 2017 = $12,000.00 |
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
A-153 ASB PROVO UT US 84602-1128 (801)422-3360 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
350 CB Provo UT US 84602-4201 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): |
Proc Sys, Reac Eng & Mol Therm, IIS Special Projects |
| Primary Program Source: |
01001617DB NSF RESEARCH & RELATED ACTIVIT 01001718DB NSF RESEARCH & RELATED ACTIVIT |
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.041 |
ABSTRACT
Abstract - Hedengren, 1547110
Current process control and optimization strategies are typically divided into three major sections: base layer controls, advanced controls, and planning and scheduling. Each of these levels works at a different time scale, ranging from milliseconds to seconds for base controls, up to months or years at the planning and scheduling level. To simplify models and decrease computation time, each of these layers receives a minimal amount of information to fulfill its objective. This lack of information creates lost opportunities. For example, the scheduling program may pass a set of unreachable set points to the advanced controls, or the base controls could reach a set point through a suboptimal path due to its lack of knowledge regarding process nonlinearities that the advanced controls had incorporated.
The purpose of this EAGER project is to develop a unified architecture for control and planning/scheduling applications. This is now possible because of large cloud computing infrastructure, mobile devices with wireless communication, smart sensors that build an internet of interacting devices, and recent break-through algorithms in solver and algorithmic performance. The project is about consolidation of disparate algorithms and research areas into a unifying architecture for scheduling and control through the development of a computational and visualization tool or APP. Although tested on local and limited multi-core resources, the project framework is proposed for a new paradigm of computing including cloud-based resources.
Undergraduate students will participate on the project for course credit. Students from underrepresented groups will be recruited directly and by connecting to an existing college program that targets such students (WE @ BYU). From an educational standpoint, students will (i) gain experience in the design and implementation process for utilization of renewable resources (ii) demonstrate an appreciation for working within engineering constraints to develop a sustainable process, (iii) demonstrate effective teamwork and leadership skills as they work together on teams, and (iv) recognize the extent to which renewable resources can be applied to multiple process platforms.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
This project developed a revolutionary approach to manufacturing optimization with combined scheduling and control. The unified approach is shown to have significant economic impact on manufacturing industries. Current optimization strategies are divided into major sections that work at different time scales, ranging from milliseconds to seconds for base controls and up to weeks or months at the planning and scheduling level. To simplify models and decrease computation time, each of these layers receives a minimal amount of information to fulfill an objective. However, this lack of information creates lost opportunities. A significant outcome of this project is the GEKKO Optimization Suite for machine learning and optimal control. The Python package combines scheduling and control into a single architecture. It also includes deep learning capabilities to build, train, and optimize in support of next-generation manufacturing. The open source software is released through GitHub and is available through pip install for Python. Other specific achievements of this project include a unified control and scheduling framework that is detailed in publications and through an online course. The publications include open access and top tier journal venues. The new approach is taught in the Dynamic Optimization course that is offered annually. It is a graduate level course on the theory and applications of numerical methods for solution of time-varying systems with a focus on engineering design and real-time control applications. Concepts taught in this course include mathematical modeling, data reconciliation, machine learning, nonlinear programming, estimation, and advanced control methods such as model predictive control.
Last Modified: 12/20/2018
Modified by: John Hedengren
Please report errors in award information by writing to: awardsearch@nsf.gov.
