Role of AI in Enabling Carbon Free Energy Transition through Predictive Maintenance

Abhinav Saxena, GE Aerospace ABSTRACT: Machine Learning and Artificial Intelligence (ML/AI) have shown great success in consumer applications and have been the main drivers for growth and innovation in the past decade. Industrial applications are fast catching up resolving their own unique set of technical, regulatory, and scalability challenges that have limited direct transferability of ML/AI as is. Significant advancements have been made in inspection, virtual sensing, dynamic process optimization, remote monitoring and predictive maintenance. However, full end-to-end deployment with system-wide coverage and autonomy still remains an elusive goal in industrial settings. Specifically, capabilities to safe-guard against unknown-unknowns, lack of explainability and trust tend to be the key bottlenecks. This session will illustrate various industrial AI/ML application examples and how these challenges are being progressively addressed as applied to energy industry. Our discussion will be in the context of reducing O&M costs in nuclear power plants where run-time robustness of ML models is key to remote monitoring and risk-informed predictive maintenance due to heavily regulated environment.