Vasilis Tsourapas

Parameter and State Estimation in Vehicle Roll Dynamics

In active rollover prevention systems, a real-time rollover index, which indicates the likelihood of the vehicle to roll over, is used. This paper focuses on state and parameter estimation for reliable computation of the rollover index. Two key variables that are difficult to measure and play a critical role in the rollover index are found to be the roll angle and the height of the center of gravity of the vehicle. Algorithms are developed for real-time estimation of these variables. The algorithms investigated include a sensor fusion algorithm and a nonlinear dynamic observer. The sensor fusion algorithm requires a low-frequency tilt-angle sensor, whereas the dynamic observer utilizes only a lateral accelerometer and a gyroscope. The stability of the nonlinear observer is shown using Lyapunovs indirect method. The performance of the developed algorithms is investigated using simulations and experimental tests. Experimental data confirm that the developed algorithms perform reliably in a number of different maneuvers that include constant steering, ramp steering, double lane change, and sine with dwell steering tests.

Model-Based Control of an Integrated Fuel Cell and Fuel Processor With Exhaust Heat Recirculation

In this paper, we consider the dynamic and controlled operation of an integrated natural gas fuel processor system (FPS), a proton exchange membrane fuel cell (PEM-FC), and a catalytic burner (CB). The FC provides power based on the electrochemical reaction of hydrogen. The FPS generates the hydrogen from natural gas through catalytic partial oxidation (CPOX) and the CB provides the energy for preheating the FPS inlet flows by burning any excess hydrogen from the FC exhaust. The coupling of these three systems poses a challenging optimization and control problem. Optimization is performed to generate the air and fuel flow intake setpoints to the FPS for various load levels. The optimal flow setpoints are used in a static feedforward map that ensures maximum efficiency at steady state. Linear quadratic techniques are then used to develop a controller to mitigate hydrogen starvation in the fuel cell and regulate CPOX reactor temperatures. We show in simulations that the designed observer-based feedback controller, which relies on temperature measurements of two reactors, speeds up the transient response fourfold, as compared to the baseline when the static feedforward controller is employed

Real-time estimation of roll angle and CG height for active rollover prevention applications

Roll angle and height of the center of gravity are important variables that play a critical role in the calculation of real-time rollover index for a vehicle. The rollover index predicts the real-time propensity for rollover and is used in activation of rollover prevention systems such as differential braking based stability control systems. Sensors to measure roll angle are expensive. Sensors to estimate the c.g. height of a vehicle do not exist. While the height of the center-of-gravity does not change in real-time, it does change with the number of passengers and loading of the vehicle. This paper focuses on algorithms to estimate roll angle and c.g. height. The algorithms investigated include a sensor fusion algorithm that utilizes a low frequency tilt angle sensor and a gyroscope and a dynamic observer that utilizes only a lateral accelerometer and a gyroscope. The performance of the developed algorithms is investigated using simulations and experimental tests. Experimental data confirm that the developed algorithms perform reliably in a number of different maneuvers that include constant steering, ramp steering, double lane change and sine with dwell steering tests.

Incremental Step Reference Governor for Load Conditioning of Hybrid Fuel Cell and Gas Turbine Power Plants

A hybrid solid oxide fuel cell and gas turbine (SOFC/GT) system exploits the complementary features of the two power plants, where the GT recuperates the energy in the SOFC exhaust stream and thereby boosting the overall system efficiency. Through model based transient analysis, however, it is shown that the intricate coupling dynamics make the transient load following very challenging. Power shutdown has been observed when the load is changed abruptly. In this work, a novel closed-loop reference governor controller is proposed to mitigate the shutdown phenomenon. The reference governor utilizes the region of attraction of a reduced order SOFC/GT model to determine the feasibility of applying an incremental step change, subject to the constraint of no system shutdown. It is shown that with a moderate computational cost, the speed of the hybrid power system response can be improved significantly compared to the fastest conventional load filter.

Control Oriented Analysis of a Hybrid Solid Oxide Fuel Cell and Gas Turbine System

The goal of this work is to investigate the feasibility of a hybrid solid oxide fuel cell (SOFC) and gas turbine (GT) system for mobile power production. A system consisting of a gas turbine, a burner, and an SOFC is examined to gain fundamental understanding of the system dynamics. A control oriented dynamic model is developed to provide the critically needed tool for system feasibility analysis and control strategy design. System optimization and transient analysis are performed based on the system model to determine the desired operating conditions and load following limitations. It is shown that the open loop system will shut down in the case of a large load step. Based on the insights learned from the open loop analysis, a feedback control scheme is proposed. The feedback scheme is based on a reference governor, which modifies the load applied to the generator to guarantee stability and fast tracking during transients.

Optimal control of hybrid electric vehicles with power split and torque split strategies: A comparative case study

In designing control strategies to optimize fuel consumption, driveability and other objectives for hybrid electric vehicles (HEVs), one can choose to use either power split or torque split as one of the control variables. While both approaches have been employed and documented, no systematic study has been reported that illuminates what implications this choice might have in terms of HEV performance, system robustness, and control strategy design and implementation complexity. This work aims to develop a case study that explores this degree of design freedom and to quantify any differences that this control design selection might impart on a given HEV architecture. Using a validated HEV model, we will derive optimal operating strategies using dynamic programming for two cases: one uses power split and other torque split. Performance metrics of fuel consumption as well as the computational complexity associated with the two different strategies will be assessed.

Incremental step reference governor for load conditioning of hybrid Fuel Cell and Gas Turbine power plants

A hybrid solid oxide fuel cell and gas turbine (SOFC/GT) system exploits the complementary features of the two power plants, where the GT recuperates the energy in the SOFC exhaust stream and thereby boosting the overall system efficiency. Through model based transient analysis, however, it is shown that the intricate coupling dynamics make the transient load following very challenging. Power shutdown has been observed when the load is changed abruptly. In this work, a novel closed-loop reference governor controller is proposed to mitigate the shutdown phenomenon. The reference governor utilizes the region of attraction of a reduced order SOFC/GT model to determine the feasibility of applying an incremental step change, subject to the constraint of no system shutdown. It is shown that with a moderate computational cost, the speed of the hybrid power system response can be improved significantly compared to the fastest conventional load filter.

A Dynamic Model of a Shipboard PEM Fuel Cell Reformer System With an Integrated Gas Turbine

A dynamic model is presented that predicts the transient characteristics of an integrated autothermal fuel reformer and gas turbine engine for extracting a high purity stream of hydrogen from logistical fuels such as F76 marine diesel or JP-5 for use in a shipboard proton exchange membrane (PEM) fuel cell power plant. The model incorporates a water-gas shift reactor for increasing hydrogen yield, and a separation membrane for extracting a pure stream of hydrogen gas from the reformer output. Compressed air supply and energy recovery is achieved by an integrated gas turbine generator. The dynamic model serves as a testing ground for the development of control methodologies and to predict limitations in transient response during electrical load variations.Copyright

Performance Evaluation of a Thermally Integrated Fuel Cell System in the Presence of Uncertainties

In this work, we focus on robustness analysis of an integrated fuel cell and fuel reforming (FCFR) system, which relies on a feedback controller to mitigate hydrogen starvation and temperature overshoot during load transitions. The fuel reformer is used to process natural gas into a hydrogen rich flow to be utilized in a proton exchange membrane fuel cell (PEM-FC). The feedback controller uses the catalytic burner (CB) and the catalytic partial oxidizer (CPOX) temperatures as measurements and adjusts the air and fuel actuator commands to assure fast load following and high steady state efficiency. Several uncertainty sources which can potentially lead to closed loop performance deterioration are considered, including CPOX clogging, hydrodesulphurizer (HDS) clogging, fuel uncertainty and CB parameter uncertainty. Steady state and transient performance are analyzed for the different uncertainty scenarios, for both open and closed loop operation (i.e., with and without feedback control). The robustness of load following and CPOX temperature regulation of the closed loop system (feedforward and feedback controlled) is established, while the open loop system (feedforward controlled) is shown to be vulnerable to all sources of uncertainties considered.