Machine learning - WG 2021-4
MACHINE LEARNING, AI, AR AND ADVANCED ANALYTICS FOR SMART ADMS
Background
Machine learning, AI, AR and Advanced Analytics are some of the most frequently used buzzwords associated with Advanced Distribution Management Solutions (ADMSs). These digital technologies will make an important contribution to address the several challenges that DSOs will face, such as the integration of renewable energies, decarbonization, e-mobility, skill renewal, safety, customer relation, asset optimization, equipment maintenance, etc.
DSO road to digital is also a reality to foster
1. Enhanced observability using New Generation of IoT (NG-IoT);
2. Distributed smart protection & control;
3. Condition & Risk Based Maintenance using:
• Analytical models, based on asset data, to predict the behaviour of assets over time and support optimal maintenance process decisions;
• Digital Twin in asset Lifecycle, which allows creation of digital replica of technical asset to pave the way to data acquisition and their conversion into knowledge.
4. Network’s operation and maintenance efficiency (e.g., Smart Alarms, WFM, NG-IoT).
5. Network’s smart management functions (e.g., Smart Alarms, ADMS, EMI, etc.);
Scope
The Working Group will cover:
• Augmented Reality (AR) solutions identification to improve field operation (MWM - Mobile Workforce Management e WFM - Work Force Management) and training solutions to increase network’s operation and maintenance efficiency;
• Assess Machine learning, Artificial Intelligence (AI) and Advanced Analytics to improve:
- network management functions;
- asset condition monitoring and health evaluation;
- adopting predictive models in terms of modernization (replacement) and maintenance strategies, assuming an ownership strategy more focused on reliability, cost and risk;
- predictive models, assessing which are the best methodologies and sensing requirements;
• To Identify asset Digital Twin performance with real impact on asset management, using Machine learning, AI, AR and Advanced Analytics
• Benchmarking of new and existing demonstration projects using Machine learning, AI, AR and Advanced Analytics
Convener :
Debusschere Vincent G2ELab - Grenoble INP France
Members :
Almoataz Youssef Abdelaziz, Ain Shams University, Egypt
Brito Mendes Patrick, E-REDES, Portugal
Cândido Carlos, E-REDES, Portugal
Cho Sung-Min Cho, KEPCO Research Institute, Korea
Davidov Sreten, Electro Ljubljana d.d, Slovenia
Duckheim Matthias, Siemens AG, Germany
El Bermawy Heba Mohammed Beder, North Delta Electricity Distribution Company, Egypt
Fereidunian Alireza, K.N. Toosi University of Technology, Iran
Gopp David, Omicron Electronics GmbH, Austria
Hany Mohamed Hasanien, Ain Shams University, Egypt
Lenz Lukas, Stromnetz Hamburg, Germany
Menzel Johannes, Schneider Electric, France
Messner Michael, Energie Steiermark Technik GmbH, Austria
Messner Michael, Energie Steiermark Technik GmbH, Austria
Pilaud Thomas, Enedis, France
Vergara Pedro P., TU Delft, Netherlands
Vergara Pedro P., TU Delft, Netherlands
Winkelmann Erik, Dekra, Germany
Won Dongjun, Inha University, Korea
Yoon Sung-Guk, Soongsil University, Korea
Zhu Hong, State Grid Nanjing Power Supply Company, China