New frequency stability assessment based on contribution rates of wind power plants

• Developed a new FSCR index to evaluate the contributions to frequency stability from different generation units. • Analyzed the variation of the FSCR index as the penetration levels of WPPs increase. • Used multiple linear regression to determine the weights of various frequency stability supports. • The new index can be preferably used in future ancillary service markets to incentivize generation units to contribute to frequency stability. In future power systems, various generation units are expected to contribute to frequency stability. Depending on their characteristics, these units will support the frequency stability in diverse ways. Traditionally, power systems have relied on synchronous generators (SGs), ensuring sufficient inertia and providing ancillary services for frequency stability. However, with the integration of multiple wind power plants (WPPs), which have substantially different characteristics, many countries are restructuring their ancillary service markets to accommodate WPPs. This study proposes a novel frequency stability contribution rate index to analyze and compare the contribution of SGs and WPPs to frequency stability. To do so, this paper first analyzes the factors from SGs and WPPs, such as inertia, power reserves, resource conditions, and available frequency responses, and compares their roles in supporting frequency stability. The proposed index is then calculated using a multiple linear regression model that incorporates these factors. Therefore, this index provides a comprehensive evaluation of SGs and WPPs in terms of their contribution to frequency stability, offering valuable insights into their roles in supporting three key frequency stability indicators, the rate of change of frequency, frequency nadir, and settling frequency. The effectiveness of the proposed index is verified through case studies on the IEEE 39-bus system. In all cases, the coefficient of determination exceeds 95%, indicating a highly predictive regression model. Furthermore, as the penetration level of WPPs rises from 20% to 60%, the contribution weight of factors related to WPPs increases while that of SGs decreases.