Dr. Christos Georgakis is a Professor of Chemical and Biological Engineering at Tufts University where he has also been the Bernard M. Gordon Senior Faculty Fellow in Systems Engineering. He described two generalizations of the classical design of experiments (DoE) methodology, the long-standing data-driven modeling methodology of choice. The first generalization enables the design of experiments with time-varying inputs, called Design of Dynamic Experiments (DoDE). The second generalization enables the development of a dynamic response surface model (DRSM) when time-resolved measurements are available. He discussed how both advances are able to contribute significantly to the modeling, optimization, and understanding of processes for which a knowledge-driven model is not easily at hand. He also argued that such approaches can be widely used in developing reduced-size meta-models, for online use in existing processes.

To view more details or a video of the lecture, please visit http://advancedsystems.engineering.uconn.edu/events/lec-lib/dll2016/.

Dr. Quan Long from United Technologies Research Center provided an overview of his recent research on Efficient Bayesian Optimal Experimental Design for Physical Models. Optimal experimental design is the key to improve data quality in engineering. Its application on real problems lags behind mainly due to the involved computational costs. Dr. Long has developed a series of methods to accelerate the computations of the utility function (expected information gain) under rigorous error control. Specifically, he has extended the applicable domain of Laplace methods from the asymptotic posterior Gaussianity, to where the shape of the posterior is characterized by noninformative manifolds. While Laplace methods require a concentration of measure, multi-level Monte Carlo method can be used to efficiently compute the nested integral of the expected information gain with a reduction of the computational complexity, even when the randomness of data dominants the shape of the posterior distribution. The developed methodologies have been applied to various engineering problems, e.g., impedance tomography, seismic source inversion and parameter inference of combustion kinetics.

To view more details or a video of the seminar, please visit http://advancedsystems.engineering.uconn.edu/events/sem-lib/sl2016/.

Dr. Bollas is scheduled to present work of his UTAS-sponsored project on Built-In Test Design for Fault Detection and Isolation of Aircraft Environmental Control Systems in the 11th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems (DYCOPS-CAB 2016), to be held in Trondheim, Norway, June 6-8, 2016. He will also co-chair the session “Performance and Fault Monitoring I” and chair the session “Process Optimization and Plantwide Control I.”

More information is available on the DYCOPS webpage: http://www.dycops2016.org/

Professor Ionnis Krevikidis spoke about data, manifold learning, and the modeling of complex/multiscale systems on March 10, 2016 as a part of the Distinguished Lecture series at the UTC Institute for Advanced Systems Engineering. He discussed some recent developments on the connection between data mining/machine learning on the one hand, and the modeling of complex/multi-scale problems on the other. The talk addressed the interface between fine scale, atomistic/stochastic codes and coarse-grained, macroscopic descriptions. In particular, Professor Krevikidis discussed (a) the reduction of stochastic simulations through diffusion maps and the use of the Mahalanobis distance, and issues of heterogeneous data fusion; (b) the issue of extending diffusion-map based simulations to new configurations/conditions; and (c) the issue of not only reducing the number of independent variables, but also reducing the number of independent parameters by taking advantage of data-mining tools.

Professor Harrison Kim from the University of Illinois at Urbana-Champaign presented a seminar on March 8, 2016 on the topic of “Complex Systems Analytics: a Promising Enabler for Sustainable Design and Manufacturing.” Designing large-scale, complex systems has been a challenging task, particularly in the predictive context of life cycle. Key challenges arise in various stages of system’s life cycle – pre-life, usage life, and end-of-life – where massive-scale data is generated and captured from complex systems design, operations, and disposal. Green Profit Design – a new term coined by Kim’s team – shows that there is a strong link between sustainable product design, user generated contents in the social network service, and corporate profit generation. Green Profit Design has been shown to be successful in designing optimal, sustainable product portfolio by use of engineering design optimization and knowledge discovery for user preference capture. In this presentation, Professor Kim presented a summary of the recent findings that there exists an optimal design and remanufacturing threshold for maximum benefit of profit and environmental impact savings. The projects are sponsored by the National Science Foundation and Deere and Co. – green, sustainable design and recovery; sustainable product family design and recovery; trend mining design for product portfolio optimization.

The title of Richard Poisson’s seminar on February 25, 2016 was Fault Detection and Isolation in Aerospace Applications – Understanding and Determining Technology Gaps. Richard Poisson discussed why faults and fault isolation are of interest to the aerospace community. He provided an overview of the current analysis methodology that is employed during the design phase for fault detection, classification and isolation. This overview was followed by a discussion of the latest products and systems in aerospace industry that are more electric, more complex, and more intelligent. This complexity and increasing reliance on computational approaches has a direct impact on fault detection and isolation. Several aspects of this emerging area of fault detection and isolation were discussed including the currently employed BIT and BITE concepts. Rich concluded the talk by identifying areas of future research in this field.