"Collaborative Proposal" on Optimal-Control Strategies Based on Comprehensive Modeling and System-Interaction Analyses for Energy-Efficient and Reduced-Emission Fuel-Cell-Energy-Systems: Power Electronics SubsystemsEPA Grant Number: R831581
Title: "Collaborative Proposal" on Optimal-Control Strategies Based on Comprehensive Modeling and System-Interaction Analyses for Energy-Efficient and Reduced-Emission Fuel-Cell-Energy-Systems: Power Electronics Subsystems
Investigators: Mazumder, Sudip K. , Nelson, Douglas , Leo, Donald J. , von Spakovsky, Michael R. , McIntyre, Chuck , Herbison, Dan
Institution: University of Illinois at Chicago , Synopsys Inc. , Virginia Polytechnic Institute and State University
Current Institution: University of Illinois at Chicago , Synopsys Inc. , University of California - Davis , Virginia Polytechnic Institute and State University
EPA Project Officer: Klieforth, Barbara I
Project Period: February 1, 2004 through January 31, 2007 (Extended to December 31, 2007)
Project Amount: $125,000
RFA: Technology for a Sustainable Environment (2003) RFA Text | Recipients Lists
Research Category: Nanotechnology , Pollution Prevention/Sustainable Development , Sustainability
The significant advantages which fuel cell based systems (FCSs) exhibit over conventional systems both in terms of energy savings and pollutant emissions depend greatly on whether or not the FCSs are well synthesized and designed and whether or not appropriate control strategies have been developed to meet the varying load profiles which these systems see in a variety of applications. To adequately address the latter both scientifically and technologically, this multi-university and multi-disciplinary collaborative proposal envisions a three-phase project in which optimal-control strategies, based on comprehensive modeling and system-interaction analyses, will be developed for the energy-efficient and reduced-emission FCSs used in hybrid stationary and non-stationary applications. In the first phase, fully transient nonlinear, general models of PEMFC stacks, different power-electronic subsystems, and a variety of balance-of-plant-subsystem components will be developed and implemented based on first principles. System/subsystem interactions and the impact of parametric and application-load variations will be investigated and resolved. The second phase will entail a bottom-up approach for formulating realistic load-profile scenarios for various types of hybrid stationary and non-stationary applications in order to generate optimal-control strategies for PEMFC systems which optimally balance overall system efficiency, cost, emissions, and reliability. In the third phase, a top-down approach will be used which utilizes multi-agent-control architecture in a multi-objective and game-theoretic framework to generate control strategies validated by the control predictions obtained using the bottom-up approach. The work just outlined entails a number of scientific and technological objectives which must be achieved by Virginia Tech (VT) in this collaborative project. Those for which the University of Illinois at Chicago (UIC) will be directly responsible or for which UIC will work in collaboration with and support of VT can be summarized as follows:
• Implementation of the PES model on the SABER1 simulator (which has been developed by one of our industrial partners, Synopsys, Inc). Modules developed in the gPROMS environment will be seamlessly integrated with SABER using iSIGHT software (distributed by Engineous Software). The resulting dynamic system model will be used to investigate the system-interaction issues and their impact on the performance, efficiency, emissions, and reliability of PEMFCs and the dynamics and stability of an overall PEMFC system due to load and parametric variations.
• Collaborate with VT in validating the control predictions developed by using the top-down approach with those obtained using VT's bottom-up approach.
• Collaborate with VT to implement the bottom-up optimal-control strategies and support VT in implementing the top-down optimal-control strategies on scaled stationary and non-stationary test-beds.
• Collaborate with VT in assessing the correct balance of off-line optimization and real-time feedback of the fuel-cell system and in determining the efficacy of the execution of the proposed schemes on real-time simulators.
During the process of control-strategy formulation, the correct balance of off-line optimization and real-time feedback for ensuring proper output tracking of the FCS will be determined while both bottom-up and top-down optimal-control strategies will be implemented on scaled stationary and non-stationary test-beds using off-line simulators (as outlined in the fourth objective above). The efficacy of the execution of the proposed schemes on real-time simulators will also be evaluated in collaboration with VT. Our mutual objectives are to integrate the virtual test-bed modeling and design using a real-time multi-channel, multi-input, and multi-output digital-signal-processing (dSPACE) platform, with the models centered on an object-oriented software structure, useful for real-time implementation through hardware-in-the-loop simulations. Optimal input profiles generated from off-line optimization will be incorporated into the control system using MATLAB functions. Real-time feedback parameters will be directly downloaded to the dSPACE digital signal processor while parameters will be adapted in real-time through the Control Desk interface.
A comprehensive modeling and understanding of the dynamics of interactive PEMFC power systems will lead to optimal controllers, which optimally balance overall system efficiency, cost, emissions, and reliability. A multi-agent intelligent optimal controller (operating in a multi-objective and game-theoretic framework) will be developed as a part of this EPA project. Such a novel controller will be adaptive and decentralized, enabling it to be flexible, application independent, and robust against disturbances and plant-parameter variations. An additional result will be an easy-to-use unified modeling framework for comprehensively modeling environmentally-friendly fuel cell power systems, which can be used for analyzing PEMFC (as well as other fuel cells) responses to variations in circuit and control parameters of the power-electronics and fluctuations in application loads. This will mitigate system-interaction problems and enable the synthesis/design of an optimal power-electronics interface for a given fuel-cell stack and application load. Furthermore, knowledge gleamed about the mechanisms of system- and component-level failures will lead to better control of the reliability of such systems.
Pollution Prevention: This research project addresses TSE research topic 2.0: Non-Reaction-Based Engineering for Pollution Avoidance and Prevention by proposing a scientific and technological investigation into breakthroughs in control systems for new and promising energy-conversion and transportation technologies, i.e., for PEM FCSs. In direct comparisons of energy savings, the PEM FCS exhibits a significant advantage over that of, for example, a conventional all-electric system (with heat pump) tied to the utility grid and a conventional all electric system, which includes an air conditioner and natural gas furnace and is tied to the utility grid. The gains both in terms of energy savings and pollutant emissions depend greatly, however, on whether or not the FCS is well synthesized and designed and whether or not appropriate control strategies have been developed to meet the varying load profiles for a variety of applications. Such significant gains in energy usage lead to a considerable reduction in CO2 emissions. Furthermore, emissions of SO2 are non-existent since the reforming of a fossil fuel to produce the H2 used by the fuel-cell stack requires that the fuel be desulphurized prior to reforming. In addition, because the reforming process occurs in a temperature range of 450°C to 850°C (depending on the fuel), the production of NOx is insignificant since most NOx formation occurs at temperatures above 1000°C. Finally, CO emissions are also insignificant since sufficient cleanup is provided to ensure that less than 10 ppm of CO is ever present in the PEMFC stack. Even small amounts of CO can poison the catalytic electrochemical process within the fuel cell itself.
1 Currently, among the commercially available simulators, SABER has the most extensive (over 30000) library of mathematical computer models in power-electronics components and systems, electromechanical-energy conversion, hydraulics, thermal, magnetic, control systems, and signal processing.