October 31, 2024
Emiliano Bellini

Researchers at GLA University in India have developed a series of harmonic mitigation techniques for islanding conditions.

“The novelty of this work lies in its unique combination of passive and active sensing strategies, optimized through adaptive control mechanisms that improve both reliability and scalability,” said lead author of the research, Sanchari Deb. pv magazine. “This approach represents a significant advance over existing methods, which tend to prioritize either accuracy or scalability, but rarely both. The proposed method is preferred for its ability to provide high detection accuracy with minimal deviation-to-zero (NDZ) notice, while maintaining system stability and power quality, making it an ideal choice for future deployments of photovoltaic systems.

Islanding occurs in distributed generation systems when there are disruptions to the power grid. Under these conditions, photovoltaic systems continue to inject the power supply to the network with negative consequences on energy quality and the risk of potential damage to electrical equipment.

“Understanding the intricacies of islanding events is paramount to designing effective solutions that ensure the seamless integration of PV systems without compromising grid performance,” the scientists said. “During islanding events in grid-connected PV systems, the presence of harmonics can exacerbate power quality issues, affecting both the grid and connected loads. »

In the newspaper “A new technique to detect and mitigate harmonics during islanding in a grid-connected photovoltaic system», published in Energy reportsThe research group studied the main factors that lead to islanding events: grid fluctuations, inverter behavior and system impedance. Their analysis aimed to analyze the specific harmonic frequencies generated by PV systems during islanding events and identify critical system parameters.

The proposed set of techniques, which the scientists described as a combination of passive and active sensing strategies, includes control algorithms and filter designs, as well as modulation strategies to suppress harmonics when operating in island. It also integrates advanced signal processing with adaptive control mechanisms. Compared to conventional approaches, it would achieve faster detection times by leveraging real-time data analysis and dynamically adapting to changing network conditions.

“Unlike existing methods, this technique is designed with scalability in mind,” the academics said. “The decentralized control structure enables seamless integration into large-scale PV networks, while adaptive algorithms ensure the system can adapt to changing network conditions in real time. »

The researchers also presented the theoretical principles of islanding detection and particularly focused on two islanding detection techniques such as sliding mode frequency shift (SMS) and phase jump detection (PJD). . They also proposed a new approach to analyze the impact of impedance mismatches on harmonic suppression during islanding.

The proposed method was validated for PV system sizes ranging from 5 kW to 100 MW and achieved a reduction in total harmonic distortion (THD) of approximately 3%, compared to 4.5% for hybrid methods. existing.

“This reduction complies with IEEE 519 standards, making the system more compliant with grid regulations while maintaining better power quality,” the scientists emphasized. “The proposed method consistently maintained an average error margin of less than 2% in detection accuracy, compared to 5-7% error in traditional methods.”

The team said selecting appropriate detection thresholds and response times for different network conditions was crucial to their work. “By optimizing these parameters, it is possible to balance the trade-off between detection speed and system stability, thereby ensuring reliable anti-islanding protection in a wide range of operating scenarios,” they concluded.

Popular content