<p>Restructuring of electricity markets coupled with deep penetration of renewable energy sources (RESs) into modern power systems has resulted into increased security concerns. The integration of wind generators into power systems significantly increases the complexity of transient stability assessment (TSA) due to the variable and intermittent nature of wind energy. The inherent fluctuations and non-linear dynamics introduced by wind generators necessitate robust modelling and real-time analysis capabilities. This research addresses the critical need for fast and accurate TSA methods specifically designed for wind generator-integrated power systems. A hybrid extreme learning machine (ELM) based ensemble method is proposed to perform TSA on detailed power systems models. The detailed models (New England 39-bus and IEEE 68-bus) are developed using the power system analysis toolbox (PSAT) software tool. They consist of 4th-order generator models with their AVRs, as well as a variable speed doubly fed induction generator (DFIG) model. The hybrid method is developed by leveraging advanced computational techniques such as the Levenberg–Marquardt backpropagation algorithm and autoencoder-based feature reduction technique. This study demonstrates that the new TSA method not only improves computational efficiency but also provides reliable stability predictions, essential for the secure operation of power grids with high levels of wind energy penetration. The findings underscore the importance of proposed stability assessment tools to manage the increased complexity and ensure the resilience of modern power systems.</p>