<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Alpisto Blog</title><description>Technical articles on engineering, scientific Python, embedded systems, and power systems.</description><link>https://alpisto.eu/</link><language>en-us</language><item><title>Deep learning for N-1 security assessment in power systems</title><link>https://alpisto.eu/blog/deep-learning-for-n-1-security/</link><guid isPermaLink="true">https://alpisto.eu/blog/deep-learning-for-n-1-security/</guid><description>Neural networks can screen contingencies 300× faster than full AC power flow — at 98%+ accuracy. A survey of CNN, GNN, and DCNN approaches, the data problem, and where the field still struggles.</description><pubDate>Sun, 24 May 2026 00:00:00 GMT</pubDate><category>power-systems</category><category>ml</category><category>deep-learning</category><category>n-1</category><category>contingency-analysis</category></item></channel></rss>