Date of Submission


Document Type


Degree Name

Master of Science in Electrical Engineering


Electrical & Computer Engineering and Computer Science


Dr. Junhui Zhao

Committee Member

Dr. Ali Golbazi

Committee Member

Dr. Mohsen Sarraf

Committee Member

Dr. Bhuiyan, Md Moinuddin


Grid-connected solar systems, Neutral point clamped (NPC) inverter, Direct power control (DPC) strategy


Phase-locked loops, Electric inverters, Solar energy, Photovoltaic power systems, Solar panels


The growing demand for clean, reliable renewable energy generation has led to the widespread adoption of solar energy as a source of electricity. Technological advancement aiding to reduce the cost of solar photovoltaic (PV) panels, as well as improvement in power electronics and control strategies for solar PV systems have also contributed to the growing popularity. For grid-connected solar systems to adequately meet future demand and grid requirements, the system must be reliable, and not affected by instability or distortions on the power grid.

In this thesis, a control strategy for single-phase grid-connected inverters that can synchronize to the grid without a phase lock loop (PLL) is proposed. The PLL is an important device that is relied on for the synchronization of solar PV systems to the electrical grid. However, the PLL has an inherently complex design and its performance is often negatively affected if the grid voltage has poor quality. In addition, eliminating the use of PLL for synchronization can avoid the issue of slow dynamic response, higher harmonics, and increased computation complexity.

The real and reactive power of the single-phase, three-level neutral point clamped (NPC) inverter is controlled by using a direct power control (DPC) strategy. A novel method of computing the power components of the single-phase inverter is proposed and this technique further improves the precision of the power components calculated by compensating the frequency and phase deviation compensation. Finally, simulations are carried out by using MATLAB/Simulink to demonstrate the effectiveness of the proposed methodology.