Marine Predator Algorithm-Based Optimal PI Controllers for LVRT Capability Enhancement of Grid-Connected PV Systems

• Published in: Biomimetics (Basel, Switzerland) Vol. 9; no. 2; p. 66
• Main Authors: Ellithy, Hazem Hassan, Hasanien, Hany M., Alharbi, Mohammed, Sobhy, Mohamed A., Taha, Adel M., Attia, Mahmoud A.
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.01.2024
• Subjects: Algorithms; Alternative energy sources; Biomimetics; Control systems; Controllers; Design; Design and construction; Electricity; Electricity distribution; Energy resources; Energy storage; Engineering research; Industrial plant emissions; low-voltage ride-through; Marine fauna; Marine Predator Algorithm; Optimization; photovoltaic; Photovoltaic power generation; Power plants; Predation (Biology); Renewable resources; Solar energy; Systems stability; Voltage; Egypt; Marine Predator Algorithm; low-voltage ride-through; photovoltaic
• DOI: 10.3390/biomimetics9020066
• ISSN: 2313-7673

Photovoltaic (PV) systems are becoming essential to our energy landscape as renewable energy sources become more widely integrated into power networks. Preserving grid stability, especially during voltage sags, is one of the significant difficulties confronting the implementation of these technologies. This attribute is referred to as low-voltage ride-through (LVRT). To overcome this issue, adopting a Proportional-Integral (PI) controller, a control system standard, is proving to be an efficient solution. This paper provides a unique algorithm-based approach of the Marine Predator Algorithm (MPA) for optimized tuning of the used PI controller, mainly focusing on inverter control, to improve the LVRT of the grid, leading to improvements in the overshoot, undershoot, settling time, and steady-state response of the system. The fitness function is optimized using the MPA to determine the settings of the PI controller. This process helps to optimally design the controllers optimally, thus improving the inverter control and performance and enhancing the system’s LVRT capability. The methodology is tested in case of a 3L-G fault. To test its validity, the proposed approach is compared with rival standard optimization-based PI controllers, namely Grey Wolf Optimization and Particle Swarm Optimization. The comparison shows that the used algorithm provides better results with a higher convergence rate with overshoot ranging from 14% to 40% less in the case of DC-Link Voltage and active power and also settling times in the case of MPA being less than PSO and GWO by 0.76 to 0.95 s.