2026 Conference Sessions

This Powergen University course is open to any attendee at an additional fee. To pre-register and reserve a seat, please visit the registration resource center or reach out to info@powergen.com. Learn more about our courses here.

COURSE#:  PGU101

COURSE DESCRIPTION:

This course introduces the fundamentals of data analytics and machine learning (ML) as applied to renewable energy systems and gas turbines. Designed to be technique-agnostic and applicable across software platforms, it provides engineers, analysts, and operations leaders with a practical foundation for building, validating, and applying analytic models in real-world energy applications.

The course begins by breaking down essential concepts—supervised vs. unsupervised learning, training vs. validation data, and model selection—before diving into four core classes of models: neural networks, clustering algorithms, classification methods, and Bayesian approaches. Participants will learn not only what these models are good for, but also where they fail.

Emphasis is placed on data preparation, cleaning, and validation, with hands-on exposure to diagnostic tools such as residual plots, confusion matrices, and ROC curves. Case studies from gas turbine operations highlight both opportunities and pitfalls, demonstrating how analytics can forecast performance shifts, detect anomalies, and classify operational states. The course also explores advanced practices including model layering, variable transformation, and managing error propagation, as well as strategies for reconciling mismatched training and validation datasets.

By the end, participants will have the tools to confidently prepare data, select and evaluate models, and communicate results effectively to technical teams and leadership.

LEARNING OBJECTIVES:

By the end of the course, participants will be able to:

• Prepare, clean, and segment performance data for use in AI/ML models.
• Select appropriate model types (neural nets, clustering, classification, Bayesian) based on objectives and constraints.
• Evaluate analytic models using diagnostic tools such as residual plots, confusion matrices, and ROC curves.
• Recognize overfitting and apply techniques to prevent it.
• Tie analytic outputs back to operational and strategic decision-making in energy systems.

Attendees should acquire:

• A working understanding of AI/ML fundamentals relevant to gas turbines and renewable energy.
• Practical tools for cleaning, structuring, and validating datasets.
• Knowledge of how to evaluate and select among different model types.
• The ability to spot common pitfalls in applying AI/ML to energy systems.
• The confidence to present model outputs in a way that builds trust with peers and leadership.

Who should attend:

• Power plant engineers, analysts and operations leaders responsible for performance monitoring and diagnostics.
• Data and digitalization professionals in utilities, IPPs, and OEMs.
• Software and modeling specialists supporting renewable energy and gas turbine operations.
• Executives and managers seeking to understand the opportunities and limitations of AI/ML in energy systems.

Prerequisite skills, knowledge, and certifications:

• No formal prerequisites required.
• Basic familiarity with gas turbine operations, renewable energy systems, or data analysis concepts is recommended.
• Suitable for professionals with limited exposure to AI/ML who want to build a functional, applied understanding.