This work proposes a novel circuit topology of the three-phase split-source inverter (SSI) and investigates its performance when fed from PV source. The proposed SSI has many advantages, such as a few passive components, continuous input DC current, can operate with the same switching states as the conventional three-phase VSI, and controls both DC link voltage and output AC voltage independently. At the onset, the single-phase SSI performance is investigated with DC-source under various loading conditions. The DC link voltage is controlled via a fixed duty cycle, and the output AC voltage is controlled using Model Predictive Control (MPC). Simulation results show that the proposed control achieves low Total Harmonic Distortion (THD) of the AC output voltage at various loading conditions. The proposed three-phase SSI performance is also investigated with DC-source, and the simulation results show that the proposed control generates AC output voltage with low THD . The proposed single- and three-phase SSI performance is then investigated when fed by a PV source. Maximum power is extracted from the PV array using PI-based incremental conductance method. Both the maximum power point tracking of the PV modules and the synthesis of the sinusoidal output load voltage are carried out within the same SSI. A bidirectional converter is used to regulate the battery. Regulation of the DC-link voltage is carried out using PI controllers . Selection of the MPPT and bidirectional PI controllers’ parameters is done using Harris Hawks optimization technique. Simulation results show that the system has succeeded in producing a constant DC-link voltage and well-regulated AC load voltage with low THD under various operating conditions. Experimental results for both single-phase and three-phase PV systems verify the system’s performance . iii For the proposed three-phase grid-connected PV SSI system two controllers are used. The first controller uses a PI-based incremental conductance method to harvest the maximum PV power, by adjusting the SSI boost switch duty cycle. In the second controller, DC-link voltage is controlled via the PI controller which generates reference currents for the MPC algorithm. The MPC uses this reference currents and employing phase lock loop (PLL) circuit to deliver the PV power to the grid with synchronized AC output current and keep the DC-link voltage regulated. Simulation results show that the system has succeeded in producing a constant DC-link voltage and well-regulated AC current with low THD under various irradiation levels.

Uninterruptible power supplies (UPS) are used in critical load applications to provide energy at power outage periods. The performance of the UPS is dependent on the quality of the output voltage. The output voltage should be sinusoidal with very low total harmonic distortion (THD), have good transient response and maintain good output voltage regulation. Various feedback control techniques that exhibit a good dynamic performance, have been proposed for the control of UPS inverters. In this work, the system performance is examined with linear quadratic optimal control. A Hamiltonian function is constructed and then a Deferential Riccati Equation (DRE) is derived to solve the LQ optimal control problem. The linear quadratic regulator (LQR) problem is then defined for both continuous and discrete time systems. The solution of the LQR problem is a state feedback gain matrix that ensures the stability of the system. The performance of the LQR is tested with single-phase UPS inverter with both linear and non-linear loads using MATLAB/SIMULINK. It is shown that the systems is able to produce sinusoidal load voltage with low harmonic content at the cost of a high surge current of the inverter output current. This, in turn, would require high power ratings of the inverter switching devices. Model predictive control (MPC) is then investigated to control the UPS inverter with a second order LC filter at its output. The advantages for using MPC are that it is easy to understand and can include multivariable cases, constraints, and nonlinearities. In particular, Finite Control-Set Model Predictive Control (FCS-MPC) is employed to control the UPS inverter and it is shown to give good overall system performance such that it meets the stringent requirement set for UPS applications. FCS-MPC is chosen since it lends itself easily to the discrete nature of the UPS power converter. The FCS- 3 MPC strategy employs the discrete model of the system to predict the future behavior of the controlled variables, and a predefined cost function is used. The power inverter switching states are selected in order to minimize the predefined cost function. The performance of the proposed control scheme using FCS-MPC is examined for single-phase voltage-source inverters with an LC- filter. The system is first examined with different types of loads (linear and non-linear loads) without constraints and then examined with constraints. MATLAB/Simulink is used to obtain the simulation results which show that the performance of the FCS-MPC can achieve a sinusoidal load voltage with low THD. An experimental verification of the FCS-MPC scheme for single phase voltage source-inverter is presented. Firstly, the experimental setup is introduced. Then, the results of both one and two step prediction horizons are presented. The results of two-step prediction horizons show that it has lower THD than that of the one-step prediction horizon.