Contingency constrained power system security assessment using cascade neural network

Abstract EN

A unified approach to power system security assessment and contingency analysis suitable for on-line applications is proposed.

The severity of the contingency is measured by two scalar Performance Indices (PIs) : Voltage-reactive power performance index, PIVQ and line MVA performance index, PIMVA.

In this paper, a two stage cascade neural network is developed : Stage I employs Multi-Layer Perceptions (MLP) neural network trained by back propagation algorithm for estimating PIs and Stage II utilizes Kohonen’s Self Organizing Feature Map (KSOFM) for contingency screening and ranking.

The effectiveness of proposed methodology is tested on IEEE 39-bus New England system at different loading conditions corresponding to single line outage.

The overall accuracy of the test results highlights the suitability of the approach for on-line applications to fast and accurate security assessment and contingency analysis.