BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260410T213238Z UID:018463D2-E28B-46B7-8C29-13AE831DD750 DTSTART;TZID=America/New_York:20260428T110000 DTEND;TZID=America/New_York:20260428T130000 DESCRIPTION:The modern electrical grid faces rapid\, cyber-enabled threats that outpace traditional defenses. Operators need immediate command filter ing and predictive assessment. However\, current cybersecurity and power-s ecurity solutions cannot stop malicious commands in time or evaluate conti ngency analysis quickly enough. This work details a unified ``intervention al security'' framework that defines a proactive cyber-physical security p aradigm that combines the ability to intercept communications (deep networ k intervention) with centralized decision-making that determines when\, wh ere\, and to what extent to intervene\, thereby mitigating unsafe states a nd trajectories before they can cause harm (interventional security). This is accomplished by employment of (i) semantics-aware in-line devices that intercept and conditionally block hazardous DNP3/TCP commands\, (ii) a de ep-neural-network power-security estimator to predict the impact of contro l actions\, and (iii) conditional human-in-the-loop decisions for final au thorization or rejection of commands. Evaluation on real utility traffic a nd with large-scale electrical transmission simulation shows sub-milliseco nd latency for packet interception and analysis\, 10-times lower error and 30-times faster evaluation than existing power-security analysis methods\ , and validation of this pipeline's ability for improving operator situati onal awareness and control. Results demonstrate this solution's ability to mitigate effects of real-world attacks like those against the Ukrainian e lectrical grid\, preventing widespread disruption before damage occurs.\n\ nSpeaker(s): Adam King\, Trevor Lewis\n\nAgenda: \n11:00am - 11:15am - Sig n in and Networking\n\n11:15am - 12:45pm - Lunch and Presentation\n\n12:45 pm - 1:00pm - Questions and Adjourn\n\n430 10th Street\, Atlanta\, Georgia \, United States\, 30332 LOCATION:430 10th Street\, Atlanta\, Georgia\, United States\, 30332 ORGANIZER:larhonda.julien@gatransmission.com SEQUENCE:33 SUMMARY:Interventional Security: Unified End-to-End Cyber-Physical Defense Leveraging Deep Network Intervention for Attack-Mitigation in Power-Contro l Networks URL;VALUE=URI:https://events.vtools.ieee.org/m/546615 X-ALT-DESC:Description: <br /><p>The modern electrical grid faces rapid\, c yber-enabled threats that outpace traditional defenses. Operators need imm ediate command filtering and predictive assessment. However\, current cybe rsecurity and power-security solutions cannot stop malicious commands in t ime or evaluate contingency analysis quickly enough. This work details a u nified ``interventional security'' framework that defines a proactive cybe r-physical security paradigm that combines the ability to intercept commun ications (deep network intervention) with centralized decision-making that determines when\, where\, and to what extent to intervene\, thereby mitig ating unsafe states and trajectories before they can cause harm (intervent ional security). This is accomplished by employment of (i) semantics-aware in-line devices that intercept and conditionally block hazardous DNP3/TCP commands\, (ii) a deep-neural-network power-security estimator to predict the impact of control actions\, and (iii) conditional human-in-the-loop d ecisions for final authorization or rejection of commands. Evaluation on r eal utility traffic and with large-scale electrical transmission simulatio n shows sub-millisecond latency for packet interception and analysis\, 10- times lower error and 30-times faster evaluation than existing power-secur ity analysis methods\, and validation of this pipeline's ability for impro ving operator situational awareness and control. Results demonstrate this solution's ability to mitigate effects of real-world attacks like those ag ainst the Ukrainian electrical grid\, preventing widespread disruption bef ore damage occurs.</p><br /><br />Agenda: <br /><p>11:00am - 11:15am - Sig n in and Networking</p>\n<p>11:15am - 12:45pm - Lunch and Presentation</p> \n<p>12:45pm - 1:00pm - Questions and Adjourn</p> END:VEVENT END:VCALENDAR