| In this dissertation, a proton exchange membrane (PEM) fuel cell-based distributed generation (DG) system is analyzed by modeling various units of the DG system, performing mathematical analyses, and simulation studies. The suitable control strategies are designed for the specific units of the DG system in order to achieve the desired operating performance.;First, the nonlinear state space model of a 500-W PEM fuel cell is developed by modeling an open-circuit output voltage of the PEM fuel cell, irreversible voltage losses in the PEM fuel cell, formation of a charge double-layer in the PEM fuel cell, along with a mass balance and thermodynamic energy balance in the PEM fuel cell system. The state space model is validated, and then used to study the dynamic behavior of the PEM fuel cell under different input conditions. The modeling of the PEM fuel cell is also performed using the neural network approach, and the nonlinear autoregressive moving average model of the PEM fuel cell with external inputs (NARMAX) is developed using the recurrent neural network. It is shown that the two-layer neural network with a hyperbolic tangent sigmoid function, as an activation function, in the first layer, and a pure linear function, as an activation function, in the second layer can effectively model the nonlinear dynamics of the PEM fuel cell.;The sizing of the lead-acid battery bank, which is the energy storage element required for the DG system, is performed using the model of a 12 V, 4Ah lead-acid battery. The discharge characteristics of the battery model are studied, and the model is appropriately scaled to perform the design of the battery bank. Using the battery and dc/dc boost converter model, the charging of the battery bank is simulated in MATLAB/Simulink. A sliding mode control law is designed for the dc/dc converter to control its output voltage.;The application of two different control strategies to a single-phase and three-phase inverter is analyzed. The objective of the control design is to achieve low (total harmonic distortion) THD output voltage, fast transient response, and asymptotic tracking of reference output voltage under linear and nonlinear loads minimizing the effect of harmonic frequencies.;First, the proportional-derivative-integral (PID) technique is used to design the voltage and current controller for the single-phase and three-phase inverter. The robust servomechanism problem (RSP) voltage controller and the sliding mode current controller are then designed for the single-phase and three-phase inverter. For both the techniques, the control design is performed in a discrete-time domain. The performance of the single-phase and three-phase inverter with these control strategies is analyzed and the results are shown. |
