From Reactive Maintenance to Predictive Intelligence: Applying AI and Machine Learning for Engineering Asset Failure Prediction
Artificial Intelligence (AI) and Machine Learning (ML) are driving a new era of innovation across engineering and technology. One of their most impactful applications is in the predictive maintenance of engineering assets, where data-driven insights can significantly improve reliability, availability, and operational efficiency.
Traditional maintenance strategies are typically either reactive, where maintenance is performed after a failure, or preventive, where maintenance is scheduled at fixed intervals. While these approaches are widely used, they can lead to unexpected downtime, reduced equipment performance, and unnecessary maintenance costs due to over-servicing.
Predictive Maintenance offers a smarter alternative. By leveraging AI and ML techniques, the future health and remaining useful life of equipment can be estimated, enabling maintenance to be performed only when needed and before critical failures occur.
This technical talk will introduce the fundamentals of AI and Machine Learning, including key learning paradigms and their practical applications. It will also explore emerging developments in the field, including the rise of Generative AI, and discuss how these technologies are shaping the future of engineering analytics.
The concepts will be illustrated through a real-world case study involving the failure prediction and life estimation of Insulated Gate Bipolar Transistors (IGBTs) used in wind turbine power converters. The presentation will cover the wind turbine system, the mechanisms leading to IGBT degradation, and the application of multiple AI/ML techniques for failure prediction and remaining life assessment. The session will demonstrate how modern AI-driven approaches can transform maintenance from a reactive activity into a proactive strategy, improving asset reliability while reducing operational costs.
Date and Time
Location
Hosts
Registration
Add Event to Calendar
Speakers
Dr. Atanu Talukdar of ABB Innovation Center
From Reactive Maintenance to Predictive Intelligence: Applying AI and Machine Learning for Engineering Asset Failure Pre
Artificial Intelligence (AI) and Machine Learning (ML) are driving a new era of innovation across engineering and technology. One of their most impactful applications is in the predictive maintenance of engineering assets, where data-driven insights can significantly improve reliability, availability, and operational efficiency.
Traditional maintenance strategies are typically either reactive, where maintenance is performed after a failure, or preventive, where maintenance is scheduled at fixed intervals. While these approaches are widely used, they can lead to unexpected downtime, reduced equipment performance, and unnecessary maintenance costs due to over-servicing.
Predictive Maintenance offers a smarter alternative. By leveraging AI and ML techniques, the future health and remaining useful life of equipment can be estimated, enabling maintenance to be performed only when needed and before critical failures occur.
This technical talk will introduce the fundamentals of AI and Machine Learning, including key learning paradigms and their practical applications. It will also explore emerging developments in the field, including the rise of Generative AI, and discuss how these technologies are shaping the future of engineering analytics.
The concepts will be illustrated through a real-world case study involving the failure prediction and life estimation of Insulated Gate Bipolar Transistors (IGBTs) used in wind turbine power converters. The presentation will cover the wind turbine system, the mechanisms leading to IGBT degradation, and the application of multiple AI/ML techniques for failure prediction and remaining life assessment. The session will demonstrate how modern AI-driven approaches can transform maintenance from a reactive activity into a proactive strategy, improving asset reliability while reducing operational costs.
Biography:
Dr. Atanu Talukdar is a Senior Principal Engineer in the Automation and Digital Instrumentation Division at ABB Innovation Center, Bangalore, India. He holds a Ph.D. in Fault-Tolerant Control from the Department of Electrical Engineering, Indian Institute of Technology (IIT) Kharagpur.
Prior to joining the industry, he served as a Scientist at the Integrated Test Range (ITR), DRDO, Chandipur, Odisha, India. With over 16 years of experience, Dr. Talukdar has been engaged in the development of both physics-based and machine learning-driven algorithms for predictive maintenance and asset health management of critical engineering systems across the power generation, renewable energy, transportation, and oil & gas sectors.
Throughout his career, he has been associated with leading global research and technology organizations, including GE Global Research, Samsung Research Institute, Baker Hughes, and ABB Innovation Center. His work has focused on advancing intelligent diagnostics, prognostics, and reliability solutions for complex industrial assets.
Dr. Talukdar's research contributions have resulted in seven U.S. patents and multiple peer-reviewed technical publications, reflecting his sustained commitment to innovation at the intersection of engineering, artificial intelligence, and predictive analytics.
Email:
Address:Senior Principal Engineer in the Automation and Digital Instrumentation Division, ABB Innovation Center, Bangalore, Karnataka, India