Grid-Friendly Integration of Wind Energy: A Review of Power Forecasting and Frequency Control Techniques

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Integrating renewable energy sources into power systems is crucial for achieving global decarbonization goals, with wind energy experiencing the most growth due to technological advances and cost reductions. However, large-scale wind farm integration presents challenges in balancing power generation and demand, mainly due to wind variability and the reduced system inertia from conventional generators. This review offers a comprehensive analysis of the current literature on wind power forecasting and frequency control techniques to support grid-friendly wind energy integration. It covers strategies for enhancing wind power management, focusing on forecasting models, frequency control systems, and the role of energy storage systems (ESSs). Machine learning techniques are widely used for power forecasting, with supervised machine learning (SML) being the most effective for short-term predictions. Approximately 33% of studies on wind energy forecasting utilize SML. Hybrid frequency control methods, combining various strategies with or without ESS, have emerged as the most promising for power systems with high wind penetration. In wind energy conversion systems (WECSs), inertial control combined with primary frequency control is prevalent, leveraging the kinetic energy stored in wind turbines. The review highlights a trend toward combining fast frequency response and primary control, with a focus on forecasting methods for frequency regulation in WECS. These findings emphasize the ongoing need for advanced forecasting and control methods to ensure the stability and reliability of future power grids.

This publication has 86 references indexed in Scilit: