Neural Network Based Decentralized Controls of Large Scale Power Systems
This document has been relocated to http://scholarsmine.mst.edu/ele_comeng_facwork/1759
There were 8 downloads as of 28 Jun 2016.
This paper presents a suite of neural network (NN) based decentralized controller designs for large scale power systems' generators, one is for the excitation control and the other is for the steam valve control. Though the control inputs are calculated using local signals, the transient and overall system stability can be guaranteed. NNs are used to approximate the unknown and/or imprecise dynamics of the local power system dynamics and the inter-connection terms, thus the requirements for exact system parameters are relaxed. Simulation studies with a three-machine power system demonstrate the effectiveness of the proposed controller designs.