Systems Approach on Advanced Utilization and Exploration of Dynamic Models of Thermal Fluids Applications
Graduate Fellows
Project Duration
2014 - 2018
Project Summary
Model-based, active fault detection and isolation (FDI) methods are developed and applied on UTC aerospace systems and open literature benchmark problems. These methods design built-in tests (BIT) that are optimal for identification and isolation of faults, detected in the form of deviation of sensed information from modeled system trajectories. This model-based FDI framework improves fault identifiability and reduces false alarm rates during maintenance testing. It is successful in dealing with uncertainty in the system boundaries, parameters and model, which resulted in substantial improvements in the relationship between false alarm and non-detection rates. Global optimization methods are developed for the minmax problems resulting from the realization of FDI as a worst-case-scenario design problem. Information stewardship through system sensitivity and Fisher Information Matrix manipulation was optimized for sensor selection in systems with multiple uncertain and noisy measurements.
Program Applications
-Initial: 787-9/10 CACTCS secondary heat exchanger
-Interim 787 Flow Sensor BIT
-Final: 787-9/10 CACTCS, 777 ECS
Project Outputs
-3 PhD and 4 undergraduate students trained
-10+ collaborators from UTAS
-3 UTAS program applications
-7 internship positions
-1 Sabbatical Lave
-1 joint patent
-13 peer-reviewed scientific papers
-14 conference presentations, 6 invited talks
Business Unit Benefit
-Structured BIT design methodology applicable to a wide range of components, subsystems and systems
-Enabling of new or improvement of existing FDI capability for aircraft systems, using model-based approach and variable/parameter inference
-Optimization-based method provides theoretical metrics for BIT capability
Tangible Benefits
-15% improvement in dispatch reliability
-Reduce false alarms and NFFs by 50%
Talent Creation
-7 Students worked on this UTAS project
-Internships: William Hale (UTAS, 2014/16), Anh Nguyen (P&W, 2014), Kyle Palmer (UTAS, 2015/16), Kyle Such (P&W, 2015/16)
-Hiring: Lu Han (job offer - UTAS/UTSCE), Anh Nguyen (hired by Modelon), Kyle Such (P&W 2016) George Bollas (Sabbatical in UTAS 2016)
Excellence Artifacts
13 scientific papers, 13 conference presentations, 6 invited lectures
IP Products
1 joint US patent application
Curriculum Enhancement
SE 5101/5201, SE 5203, SE 50102
Business Unit Engagement
10+ collaborators from UTAS, 200 hrs (4 yrs) of UTAS engagement
Example Publications and Further Reading
-Palmer KA, Hale WT, Bollas GM. Optimal Design of tests for heat exchanger fouling identification, Appl. Thermal Eng. 2016;95:382-93.
-Palmer KA, Hale WT, Bollas GM. Built-in Test Design for Fault Detection and Isolation in an Aircraft Environmental Control System, IFAC Papers Online. 2016;49(7):7-12.
-Bollas GM, Palmer K, Prasad D, Park Y, Maljanian J, Poisson R, Jacobson C. Plate-fin Heat Exchanger Fouling Identification. US Patent Application #201562172486, EU Patent Application #16173593.1 - University of Connecticut and United Technologies Corporation, 2016.
Peer Reviewed Papers
Palmer KA, Hale WT, Bollas GM. Optimal Design of tests for heat exchanger fouling identification, Appl. Thermal Eng. 2016;95:382-93.
Han L, Zhou Z, Bollas GM. Model-based analysis of chemical-looping combustion experiments. Part II: Optimal design of reduction experiments. AIChE J. 2016;62(7):2419-31.
Han L, Zhou Z, Bollas GM. Model-based analysis of chemical-looping combustion experiments. Part I: Structural identifiability of kinetic models. AIChE J. 2016;62(7):2432-46.
Han L, Bollas GM. Dynamic optimization of fixed bed chemical-looping combustion systems integrated in thermal power plants. IFAC Papers Online. 2016;49(7):115-20.
Palmer KA, Hale WT, Bollas GM. Built-in Test Design for Fault Detection and Isolation in an Aircraft Environmental Control System, IFAC Papers Online. 2016;49(7):7-12.
Palmer KA, Bollas GM. Active Fault Detection and Identification using Transient Data. Computer Aided Chemical Engineering. 2017;40:1687-1692.
Papers in Review
Hale WT, Palmer KA, Bollas GM. Discrete fault identification and isolation in complex systems. IEEE Access. 2018; in review.
Palmer KA, Hale WT, Bollas GM. Active fault identification by optimization of test Designs, IEEE TAC. 2018; in review.
Halle WT, Stuber DM, Bollas GM. Robust design of built-in tests for active fault detection and isolation of discrete faults in uncertain systems. Chemistry, Molecular Sciences and Chemical Engineering – Process and Systems Engineering – System Identification Chapter. 2018; invited paper - in preparation.
Palmer KA, Bollas GM. Active fault diagnosis with optimal test designs for open-loop uncertain systems. Automatica. 2018; in review.
Palmer KA, Hale WT, Bollas GM. Active Fault Diagnosis with Sensor Selection in a Diesel Engine Air Handling System, proceedings of the American Control Conference. 2018; in review.
Hale WT, Stuber DM, Bollas GM. Design of Built-In Tests for Robust Active Fault Detection and Isolation of Discrete Faults in Uncertain Systems, proceedings of the American Control Conference. 2018; in review.
Palmer KA, Bollas GM. Analysis of Transient Data in Test Designs for Active Fault Detection and Identification. Computers and Chemical Engineering. 2018; invited paper - in review.
Conference Presentations
Palmer KA, Bollas GM. Optimal test design framework for model-based active fault detection and isolation. In: 2018 AIChE Annual Meeting. Minneapolis, MN, USA, 2017.
Palmer KA, Bollas GM. Active Fault Detection and Identification using Transient Data. In: European Symposium on Computer Aided Process Engineering, ESCAPE-27. Barcelona, Spain, 2017.
Hale W, Palmer K, Bollas GM. Mixed-integer non-linear programs for optimal fault detection and isolation procedures. In: 2017 FOCAPO/CPC Conference, Tucson, AZ, USA, 2017.
Palmer KA, Hale WT, Bollas GM. Active Fault Detection and Isolation and False Alarm Elimination by Constrained Optimization of Built-In and Maintenance Test Conditions. In: 2016 AIChE Annual Meeting, San Francisco, CA, USA, 2016.
Han L, Bollas GM. A model-based framework for dynamic optimization in power generation systems. In: 11th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems (DYCOPS-CAB 2016), Trondheim, Norway, 2016.
Palmer KA, Hale WT, Han L, Jacobson CA, Bollas GM. Built-in Test Design for Fault Detection and Isolation in an Aircraft Environmental Control System. In: 11th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems (DYCOPS-CAB 2016), Trondheim, Norway, 2016.
Magnusson F, Palmer K, Hale W, Han L, Bollas GM. Dynamic parametric sensitivity optimization using simultaneous discretization in Jmodelica.org. In: 1st International Conference on Complex Systems Engineering, Storrs, CT, 2015.
Hale W, Palmer K, Bollas GM. Built-In Tests for Thermal Fluid Systems of Aerospace Applications. In: 2015 AIChE Annual Meeting, Salt Lake City, UT, USA, 2015.
Palmer KA, Hale WT, Bollas GM. Plate-Fin Heat Exchanger Fouling Identification using D-Optimal Experimental Design. In: Heat Exchanger Fouling and Cleaning XI, Enfield, Ireland, 2015
Palmer K, Magnusson F, Hale W, Bollas GM. Dynamic parametric sensitivity optimization using simultaneous discretization in Jmodelica.org. In: 2015 North America Modelica Users' Group Meeting, Storrs, CT, 2015
Magnusson F, Palmer K, Hale W, Han L, Bollas GM. Dynamic parametric sensitivity optimization using simultaneous discretization in Jmodelica.org. In: 1st International Conference on Complex Systems Engineering, Storrs, CT, 2015
Palmer KA, Hale WT, Bollas GM. Plate-Fin Heat Exchanger Fouling Identification using D-Optimal Experimental Design. In: Heat Exchanger Fouling and Cleaning XI, Enfield, Ireland, 2015
Han L, Zhou Z, Bollas GM. Model discrimination and parameter precision using theories of parameter identifiability and design of experiments. In: UTC IASE 1st Annual Conference on Systems Engineering, November 2014.
Palmer K, Hale W, Han L, Bollas GM. Design of Optimal Built-In Tests via System Sensitivity Analysis. In: UTC IASE 1st Annual Conference on Systems Engineering, November 2014.
Invited Lectures (George Bollas)
“Design of Built-In Tests for Fault Detection and Isolation,” Carnegie Mellon University – Pittsburgh, PA, USA, September 2017
“Model-based systems engineering at the cross-section of smart manufacturing, smart buildings and power generation systems,” Biomedical Engineering Alliance & Consortium, North Haven (Medtronic campus), CT, USA, April 2017
“UTC Institute for Advanced Systems Engineering: Fault Detection and Isolation in Cyber-Physical Systems,” United Technologies Research Center - Ireland, June, 2016.
“Structured iBIT design for the identification of fouling in aircraft heat exchangers,” California Institute of Technology, iCyPhy meeting on Built-In Tests, January, 2015.
“Systems Approach on Advanced Utilization and Exploration of Dynamic Models of Thermal Fluids Applications,” UTC IASE 1st Annual Conference on Systems Engineering, November 2014.
“Parameter identifiability and optimal experimental design: Theory and application on the identification of reaction kinetics in chemical-looping combustion,” Lund University, Department of Automatic Control, Seminar Series, September 2014.
Patent Applications
Bollas GM, Palmer K, Prasad D, Park Y, Maljanian J, Poisson R, Jacobson C. Plate-fin Heat Exchanger Fouling Identification. US Patent Application #201562172486, EU Patent Application #16173593.1 - University of Connecticut and United Technologies Corporation, 2016.