ON-LINE POWER-SYSTEM STABILITY ASSESSMENT USING PATTERN RECOGNITION METHODSDIMITROVSKA, TEODORA (Avtor) Mihalič, Rafael (Mentor) electrical power-systemtransient stability assessmentpattern recognitionprincipal comoponents analysiseuclidean distancecontingency screening methodalgorithm for synthesisFor many utilities around the world, there has been considerable pressure to increase power flows over existing transmission corridors, partly due to economic incentives such as trends towards deregulation and competition. This consistent pressure has prompted the requirement for extending common power-system studies to take account of dynamic security assessment and control. Furthermore, several post-mortem analyses have shown that scenarios that have had catastrophic impacts such as blackouts are rarely caused by an isolated single or multiple common-mode contingencies that are to blame but a handful of independent contingencies coupled with an already stressed system operating state. The lack of situational awareness has been identified as one of the main factors contributing to the gravity of these events. Moreover, network planing and developement trends are expected to introduce an unprecedented level of uncertainty to the system operators, not only in the locations of distributed resources and active loads, but their intermittent nature as well. The so-called tribal knowledge of system operators is failing in dealing with completely unfamiliar patterns of the system behaviour. This is why it is important to understand and document both the current and future patterns of system behaviour with regards to dynamic stability.20192019-06-12 13:45:10Doktorsko delo/naloga108034sl