Enhancement of power system transient stability using machine learning and MPC

  • Abdula Fawzy Saleem

Student thesis: Master's Thesis

Abstract

This thesis report introduces a new approach to identify generator coherency for the use of damping control of inter area modes of oscillations in power systems. The low frequency oscillations become present after any type of disturbances in the network, which is due to the imbalance in electrical and mechanical torque in the machines. This consequently limits the capabilities of the network to transmit power as well as degrade the stability of the grid. This work proposes to use the concept of enhanced generator coherency identification using machine learning to aid the damping control of such oscillations using the MPC algorithm for the implementation of a wide-area damping control (WADC) scheme. The introduced coherency identification technique is based on hierarchical clustering in addition to a hybrid method for clustering based on distance and link inconsistency. The WADC scheme incorporates modal analysis, residual analysis, parameter estimation, state estimation as well as MPC technique for damping control.
Date of AwardJul 2021
Original languageAmerican English

Keywords

  • Wide Area Damping Control (WADC)
  • Generator Coherency Identification (GCI)
  • Transient Stability Identification (TSI)
  • Model Predictive Control (MPC).

Cite this

'