Low Voltage Ride-through for Photovoltaic Systems Using Finite Control-Set Model Predictive Control

• Other Authors: Franco, Fernand Diaz (authoraut)
• Format: Others
• Language: English; English
• Published: Florida State University
• Subjects: Electrical engineering
• Online Access: http://purl.flvc.org/fsu/fd/FSU_SUMMER2017_DiazFranco_fsu_0071E_14045

Grid codes impose immunity requirements to the generation systems that are connected to the transmission lines. Immunity refers to the generator’s capability to overcome grid abnormal conditions. One of the requirements is to remain connected during a certain time when a fault, like voltage sag, is presented. During the fault scenario, a generator unit should remain connected for a pre-determined amount of time, and also provide reactive power to support the grid voltage. This is called low-voltage ride through (LVRT). Initially, LVRT requirements were imposed for large generator units like wind farms connected to the transmission network; however, due to the increased penetration of distributed generation (DG) on the distribution system, new grid codes extend the mentioned capability to generator units connected to the distribution grid. Due to matured photovoltaic (PV) technology and the decreased price of PV panels, PV grid tied installations are proliferating in the utility grids; this is creating new challenges related to voltage control. In the past, DG such as PV were allowed to trip from the grid when a fault or unbalance occurred and reconnect within several seconds (sometimes minutes) once the fault had been cleared. Nevertheless, thanks to high PV penetration nowadays, the same method cannot be used because it will further deteriorate the power quality and potentially end in a power blackout. Different approaches have been considered to fulfill the LVRT requirement on PV systems. A large amount of literature focuses on the control of the grid side converter of the PV installation rather than the control of PV operation during the fault, and most control designs applied to the grid side follow classical control methods. Moreover, the effects of the grid fault on the generator side impose a challenge for controlling the PV systems since the quality of the synthesized converter voltages and currents depends on the dc link power/voltage control. This document proposes a Model based Predictive Control (MPC) for controlling a two stage PV system to fulfill LVRT requirements. MPC offers important advantages over traditional linear control strategies since the MPC cost function can include constraints that are difficult to achieve in classical control. Special attention is given to implementation of the proposed control algorithms. Simplified MPC algorithms that do not compromise the converter performance and immunity requirement are discussed. === A Dissertation submitted to the Department of Electrical and Computer Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. === Summer Semester 2017. === July 21, 2017. === LVRT, MPC, Photovoltaic, Renewable Energies, Voltage support === Includes bibliographical references. === Chris S. Edrington, Professor Directing Dissertation; Juan Ordonez, University Representative; Omar Faruque, Committee Member; Simon Y. Foo, Committee Member.