A5091115429- Advanced Control Systems Optimization
- Control Systems and Identification
- Fault Detection and Control Systems
- Numerical Methods and Algorithms
- Advanced Optimization Algorithms Research
- Probabilistic and Robust Engineering Design
- Formal Methods in Verification
- Numerical methods for differential equations
- Polynomial and algebraic computation
- Risk and Portfolio Optimization
- Carbon Dioxide Capture Technologies
- Model Reduction and Neural Networks
- Optimization and Variational Analysis
- Membrane-based Ion Separation Techniques
- Membrane Separation Technologies
- Optimal Power Flow Distribution
- Ammonia Synthesis and Nitrogen Reduction
- ERP Systems Implementation and Impact
- Reservoir Engineering and Simulation Methods
- Thermodynamic and Exergetic Analyses of Power and Cooling Systems
- Electric Power System Optimization
- Metal-Organic Frameworks: Synthesis and Applications
- Reliability and Maintenance Optimization
- Extremum Seeking Control Systems
- Power System Reliability and Maintenance
Georgia Institute of Technology
2020-2024
University of Waterloo
2024
University Hospital Southampton NHS Foundation Trust
2024
Clemson University
2014-2020
University of Wisconsin–La Crosse
2018
Massachusetts Institute of Technology
2009-2015
IIT@MIT
2009-2010
Membrane-based organic solvent separations are rapidly emerging as a promising class of technologies for enhancing the energy efficiency existing separation and purification systems. Polymeric membranes have shown promise in fractionation or splitting complex mixtures molecules such crude oil. Determining performance polymer membrane when challenged with mixture has thus far occurred an ad hoc manner, methods to predict based on composition chemistry unavailable. Here, we combine...
A robust control method is presented for linear systems subject to input and state constraints, bounded disturbances measurement noise, discrete faults in sensors, actuators, system dynamics. The approach uses set-based fault detection isolation techniques coordinate switching between controllers designed each scenario. In contrast previous approaches, the an active guarantee isolability constraints.
SUMMARY Convex and concave relaxations for the parametric solutions of ordinary differential equations (ODEs) are central to deterministic global optimization methods nonconvex dynamic open‐loop optimal control problems with parametrization. Given a general system ODEs parameter dependence in initial conditions right‐hand sides, this work derives sufficient under which an auxiliary describes convex solutions, pointwise independent variable. Convergence results these also established. A fully...
A deterministic algorithm for solving nonconvex NLPs globally using a reduced-space approach is presented. These problems are encountered when real-world models involved as nonlinear equality constraints and the decision variables include state of system. By model equations dependent (state) implicit functions independent (decision) variables, significant reduction in dimensionality can be obtained. As result, inequality objective function which estimated via fixed-point iteration. Relying...
We propose a novel global solution algorithm for the network-constrained unit commitment problem incorporating nonlinear alternating current model of transmission network, which is nonconvex mixed-integer programming (MINLP) problem. Our based on multi-tree optimization methodology, iterates between lower-bounding and upper-bounding exploit mathematical structure with AC power flow constraints (UC-AC) leverage optimization-based bounds tightening, second-order cone relaxations, piecewise...
Effective fault diagnosis depends on the detectability of faults in measurements, which can be improved by a suitable input signal. This article presents deterministic method for computing set inputs that guarantee diagnosis, referred to as separating inputs. The process interest is described, under nominal and various faulty conditions, linear discrete-time models subject bounded measurement noise. It shown efficiently computed terms complement one or several zonotopes, depending number...
This article considers the design of an input signal for improving diagnosability faults from process measurements. Previous work has focused on open-loop design. In particular, deterministic methods are available computing that guarantees fault diagnosis within a specified time horizon, whenever such exists. Here, two closed-loop approaches considered use feedback in order to reduce length and/or cost required input, while maintaining this guarantee. The first method uses existing receding...
The problem of designing an input to improve the detectability faults has been addressed previously using both stochastic and deterministic formulations. This article presents a hybrid approach that provides worst-case guarantee diagnosis within time interval [0,N], while maximizing probability at some earlier M <; N. Compared purely methods, this strategy reduces average required for diagnosis. Moreover, high enables N be chosen large, thus reducing conservatism input.