Correction to ´Preliminary results on using artificial neural networks for security assessment´ (May 91 890-896)

The discussion and the closure which should have appeared in the above paper (see ibid., vol.6, no.2, p.890-6, May 1991) are presented. The discussers point out that the neural net approach for assessing online the dynamic security of a power system proposed by the authors is still not ready for a practical use. In their reply, the authors agree with the commenters, but emphasize that their paper is an attempt to determine whether such an approach has promise.<>