Energy security and AI BSPC 2025-05-20 v1

Slide 1:Energy security and digitalisationTarmo Korõtko, PhDSenior ResearcherDept. of Electrical Power Engineering, School of EngineeringTallinn University of TechnologyTarmo.korotko@taltech.eeSlide 2:energy securityEnergy Policy trilemma:energy securityaffordabilitysustainabilityEnergy security means ensuring a reliable, continuous, and adequate supply of energy to meet demand under various circumstancesResilient energy infrastructure is fundamental to achieving and maintaining energy securityImage generated using Sora by OpenAI (plan Plus)Slide 3:System resilience (or Resiliency)A system is resilient to the degree to which it rapidly and effectively protects its critical capabilities from harm caused by adverse events and conditions [1]Resilience assumes adversities are inevitable and focus lies on the system's responseAdverse conditions enable adverse events, which are stressful situations that can disrupt critical capabilitiesImage generated using Sora by OpenAI (plan Plus)Slide 4:Adversity TimelineSlide 5:Digitalisation and AIArtificial Intelligence (AI): intelligence exhibited by machines/softwareAI enables to process massive amounts of multidimensional and -modal data generated by modern energy systemsTo become AI-native, one must become digital firstDigital and AI-based tools provide numerous benefits (e.g. enhanced system stability and control, improved reliability and resilience, increased efficiencies, accurate forecasts, condition-based maintenance, etc.) but also increase complexity and opacity (reduced transparency)AI is not the solution to our problems, but a tool to be used in our solutionsDigitalisationSlide 6:Adverse conditions and their detectionQuality Attribute | Adverse Condition | Data-driven methods for detecting adverse conditionsRobustness | Hardware, software or data defects | Predictive maintenance, automated log analysis and data-pipeline monitoring systemsSafety | Operational hazards | Automated equipment health monitoring, access control and automated CCTV processingEnvironmental | Extreme environmental conditions | Combine real-time sensor data, historical data, satellite images and weather data to train models for detection and predictionCybersecurity | Cyber-threats | Network traffic and multi-level log data to train ML and AI models for mitigating cyber-threatsSurvivability | Kinetic or military threats | Real-time sensor data, UxV feeds, satellite imagery with computer vision and geospatial AI toolsCapacity | Overloading | Real-time sensor and measurement data and rule-based monitoring systemsLongevity | Equipment ageing | Asset management tools and predictive maintenanceInteroperability | Communication degradation | Real-time AI analysis of streaming telemetry and network logsSlide 7:Adversity timeline(1)Slide 8:Addressing system resiliencyRequirements EngineeringArchitectureTechnical DesignImplementationVerificationRisk AnalysisHarm AssessmentAsset PrioritisationHigh-level RequirementsDetermination and Prioritisation of AdversitiesRequirements for Quality AttributesInspectionAnalysisDemonstrationTestingSlide 9:Digital technologies for system resiliency testingImages generated using Sora by OpenAI (plan Plus)Slide 10:SummaryResilient energy infrastructure is a key component of energy securityAvoiding adverse conditions is the preferred measure for improving energy securityDigital technologies enable to detect and mitigate adverse conditionsImplementing digital tools increases system complexity, reducing resilienceIntegrating AI at scale requires a system with a high level of digitalisationResilient systems assume that adversities will happenRequirements engineering and Verification are key in producing resilient systemsDigital technologies enable extensive resiliency testingSlide 11:Tarmo Korõtko, PhDSenior ResearcherDept. of Electrical Power Engineering,School of Engineering,Tallinn University of TechnologyTarmo.korotko@taltech.eeThank You!Slide 12:ReferencesD. Fairsmith, “System Resilience: What Exactly is it?” Software engineering Institute, Carnegie Mellon University, 2025. Online material. Accessed 15.05.2025. url: https://insights.sei.cmu.edu/blog/system-resilience-what-exactly-is-it/M. N. Dehaghani, T. Korõtko and A. Rosin, "AI Applications for Power Quality Issues in Distribution Systems: A Systematic Review," in IEEE Access, vol. 13, pp. 18346-18365, 2025, doi: 10.1109/ACCESS.2025.3533702.Arqum Shahid, Roya Ahmadiahangar, Argo Rosin, Andrei Blinov, Tarmo Korõtko, Dmitri Vinnikov, "Leveraging the machine learning techniques for demand-side flexibility – A comprehensive review," Electric Power Systems Research, Volume 238, 2025, 111185, ISSN 0378-7796, https://doi.org/10.1016/j.epsr.2024.111185.A. Shahid, F. Plaum, T. Korõtko and A. Rosin, "AI Technologies and Their Applications in Small-Scale Electric Power Systems," in IEEE Access, vol. 12, pp. 109984-110001, 2024, doi: 10.1109/ACCESS.2024.3440067.Zahraoui, Younes, Tarmo Korõtko, Argo Rosin, Saad Mekhilef, Mehdi Seyedmahmoudian, Alex Stojcevski, and Ibrahim Alhamrouni. 2024. "AI Applications to Enhance Resilience in Power Systems and Microgrids—A Review" Sustainability 16, no. 12: 4959. https://doi.org/10.3390/su16124959Alhamrouni, Ibrahim, Nor Hidayah Abdul Kahar, Mohaned Salem, Mahmood Swadi, Younes Zahroui, Dheyaa Jasim Kadhim, Faisal A. Mohamed, and Mohammad Alhuyi Nazari. 2024. "A Comprehensive Review on the Role of Artificial Intelligence in Power System Stability, Control, and Protection: Insights and Future Directions" Applied Sciences 14, no. 14: 6214. https://doi.org/10.3390/app14146214Mischos, S., Dalagdi, E. & Vrakas, D. Intelligent energy management systems: a review. Artif Intell Rev 56, 11635–11674 (2023). https://doi.org/10.1007/s10462-023-10441-3D. Fairsmith, “System Resilience Part3: Engineering System Resilience Requirements,” Software engineering Institute, Carnegie Mellon University, 2025. Online material. Accessed 15.05.2025. url: https://insights.sei.cmu.edu/blog/system-resilience-part-3-engineering-system-resilience-requirements/