2027 Event Agenda

As utilities, system operators, and planners face unprecedented growth in hyperscale, cloud, and AI-driven data centers, accurately representing these facilities in power system studies has become a critical industry challenge. Traditional load modeling approaches were not developed to capture the unique characteristics of modern data centers, which increasingly consist of power-electronic-based loads, sophisticated UPS systems, battery energy storage, backup generation, and rapidly evolving computing technologies. As a result, utilities and planners are seeking practical methodologies for representing data center behavior in transmission planning, interconnection studies, reliability assessments, dynamic simulations, and operational analyses.

This Utility University course provides a comprehensive and practical framework for understanding and modeling modern data centers across a range of power system applications. Attendees will gain insight into how data centers are electrically designed, how their major components interact with the grid, and how various modeling approaches can be used to represent facility performance under both normal and disturbed system conditions.

The course begins with an overview of modern data center architectures, including enterprise, colocation, hyperscale, and AI/high-performance computing (HPC) facilities. Key electrical infrastructure components, including medium-voltage distribution systems, UPS systems, battery energy storage systems, standby generators, cooling systems, and IT loads, will be examined from both operational and modeling perspectives.

Participants will explore the modeling techniques currently used throughout the industry, including traditional load representations, ZIP models, WECC Composite Load Models (CMLD), CLOD models, and custom user-developed approaches. The course will discuss the strengths, limitations, assumptions, and appropriate applications of each methodology when representing modern data center loads.

The course will also examine emerging data center-specific modeling methodologies, including models developed by PNNL and ERCOT, along with K&A’s internally developed data center model. Attendees will gain insight into the objectives, architectures, assumptions, aggregation techniques, and intended applications of each approach. Practical implementation considerations, validation challenges, and lessons learned from real-world model development efforts will also be discussed.

A major focus of the course is the development and implementation of data center models for both electromagnetic transient (EMT) and planning studies. Through practical examples and PSCAD-based demonstrations, attendees will learn how key facility components are represented, how detailed models can be simplified for large-scale studies, and how aggregation techniques can be applied while maintaining sufficient modeling fidelity for the intended application. The course will discuss model development workflows, parameter selection, model assumptions, and practical considerations for ensuring that models are appropriate for their intended study objectives.

Attendees will learn how detailed EMT models developed in PSCAD can be translated into planning-level representations suitable for PSS®E studies while maintaining consistency between electromagnetic transient and RMS-domain simulations. Practical examples will be used to illustrate differences between PSCAD and PSS®E simulation results and the modeling considerations required to achieve consistent behavior across PSCAD and PSS®E.

Throughout the course, instructors will discuss current industry practices, utility expectations, reliability considerations, and emerging best practices related to data center modeling. By the end of the course, attendees will understand the current state of industry practice, recognize the strengths and limitations of existing modeling approaches, and be better equipped to evaluate, develop, and apply data center models for planning, interconnection, reliability, and operational studies.

The course is intended for utility planners, system operators, consultants, developers, manufacturers, and engineers seeking practical guidance on accurately representing one of the fastest-growing and most consequential new load types on the modern electric grid.