PMAPS 2026 Salt Lake City, UT

Abstract

This tutorial will provide an overview of the traditional challenges associated with expansion planning problems for the grid, including the combinatorial complexity of component selection and timescale dependency. It will also highlight the advantages and trade-offs of using Generalized Disjunctive Programming (GDP) as an alternative optimization paradigm compared to traditional mixed-integer formulations.

The tutorial will introduce users to the IDAES-GTEP software package, an IDAES- and Pyomo-based tool for building modular, flexible GDP models to solve power infrastructure planning problems. Participants will be shown a pre-built notebook describing how to build test cases, format input data, and interpret the outputs of the IDAES-GTEP tool to solve expansion planning problems. The second half of the tutorial will support a more in-depth and participant-driven case study adapted to audience interests.

By the end of the sessions, participants will be able to run the package, solve and analyze basic test cases, and build their own case study from scratch. Participants will learn the basics of Generalized Disjunctive Programming, how to formulate and solve models using Pyomo.GDP, how to build and customize IDAES-GTEP model components, and how to prepare input data for using IDAES-GTEP.

Abstract

Compounded extreme weather events in the electric grid occur when multiple severe weather stressors—such as heatwaves, droughts, and wildfires—coincide or interact, imposing simultaneous, correlated pressures on system adequacy and operability, leading to cascading posing reliability threat. Over recent decades, both the frequency and duration of these events have escalated, amplifying risks of prolonged outages, public safety threats, economic losses, and disruptions to critical services. Preparing the grid for such complex phenomena requires investments in grid hardening, better forecasting and monitoring, and—critically—realistic models that represent resource constraints under stress. As renewable penetration rises, hydropower’s role as a flexible, fast-ramping resource becomes pivotal; however, prevailing planning and operations models often under-represent hydropower’s dependency on water availability and operational constraints.

Objective of this tutorial is to present a comprehensive framework of how basceses (powerflow and dynamic models) are developed in the Western Interconnection, and how stress-event production cost modeling (PCM) is coupled with basecases and how accurate hydrological and weather data are then brought into the electrical model using statistical and probabilistic tools, to develop high-impact event cases used for planning studies. The other objective is to demonstrate impact of accurate representation of hydrological data on power system reliability during high impact events.

Abstract

Resilience metrics are rapidly evolving in both methods and applications, and this tutorial will describe the state of the art with an emphasis on advances in probabilistic methods, available utility data, industry needs, and emerging solutions. Resilience methods for both distribution and transmission will be treated, with more emphasis on distribution.

There has been much progress on developing detailed physical models to simulate and optimize resilience, but less emphasis until very recently on practical processing of utility data into resilience metrics and analyzing their statistics. This tutorial will focus on opportunities for processing data and will show that much can be done relatively easily with data already routinely collected by utilities. While much of these data are confidential, public data sources available to all are increasing.

Once resilience events and metrics are extracted from utility data, we can systematically quantify what actually happened in historical resilience events. This grounds and fosters further analysis and engineering solutions. Typical patterns of transmission system restoration can be characterized by a Poisson process with variable rates. However, there is significant variation in historical event sizes that manifests as heavy-tail phenomena in the distributions of resilience metrics. This is an inherent property of extreme events in power systems and is challenging because many seemingly sensible metrics become unworkable in practice.

The tutorial will explain heavy-tail phenomena and emerging metrics that address these challenges. It will also describe industry needs and experiences, including practical approaches to resilience that are emerging and shaping investments in resilience. Case studies will illustrate the methods, with the goal of coordinating advances and challenges in industry with research opportunities in resilience, particularly probabilistic analyses.

Abstract

Integration of intermittent resources, changing weather patterns and the rapid increase of interconnection requests for large-scale electricity consumers have further complexified the already challenging process of power system expansion planning. As safely accelerating the planning process becomes a more urgent need for utilities and policy makers, interest grows in tools capable of more detailed consideration of spatial resolution and of handling uncertainty and variability. These requirements translate into much larger optimization problems at the frontier of what is computationally feasible. Parallel computing is a tool to advance that frontier.

In this tutorial, we cover recent advances in large-scale capacity expansion planning, focusing on the Coordinated Resource Investment Tradeoff Toolkit for Energy Resilience (CRITTER), an open-source software tool developed at Lawrence Livermore National Laboratory and made available to the research community. Through small, interactive examples, we cover the necessary basics of power system expansion planning, stochastic optimization, and high performance computing (HPC), all of which CRITTER incorporates.

In the first session, we present examples of how to run stochastic optimization problems in parallel with the Python package mpisppy, the tool at the heart of the CRITTER toolkit. After the session, the student should be able to understand and run a textbook example case.

In the second session, we introduce the Capacity Expansion Planning (CEP) problem to illustrate the value of parallel stochastic programming tools and the challenges involved in solving real-world problems. The examples presented serve to demonstrate the difficulty of solving large-scale, stochastic, power system expansion problems as well as the benefit that increasingly powerful HPC capabilities can provide in solving them. Small illustrative examples will be solved interactively by the students. Finally, we will briefly cover the complementary power system modeling software tool EGRET, other useful features of mpisppy, and possible extensions of the CEP problem.

Abstract

The increasing penetration of distributed energy resources, electric vehicles, and flexible loads has transformed traditional distribution networks into highly dynamic Active Distribution Grids (ADGs). An active distribution network is a complex distribution system composed of loads, distributed generation, energy storage, energy conversion devices, electric vehicles, monitoring devices, and protection devices.

The new grid introduces operational uncertainties and complexities that demand robust reliability assessment techniques. Traditional reliability assessment models and methods are significantly influenced by these changes, and there are a variety of challenges in the reliability assessment of active distribution networks.

This tutorial introduces a comprehensive overview of modern methods for evaluating reliability in ADGs, covering analytical, simulation-based, and data-driven approaches, including renewable generation uncertainty and failure probability. Through practical examples and case studies, participants will gain the knowledge required to accurately assess system performance, identify critical vulnerabilities, and support resilient planning and operation of next-generation distribution networks.

Abstract

Oscillation monitoring is a critical aspect of reliable power system operation. In this tutorial, attendees will learn the theoretical underpinnings of oscillation analysis and then see what commercial solutions look like in practice. The objectives of the tutorial are to 1) enhance the ability of grid operators to effectively use their existing tools, and 2) inform researchers of practical constraints and preferences that should be considered when developing new tools.

In the first session, three oscillation analysis methods grounded in statistical theory will be discussed. These include the application of statistical theory to detect oscillations in SCADA measurements, the root mean squared (RMS)-energy method deployed by several grid operators to monitor synchrophasor measurements, and recent developments in the application of the RMS-energy method to detect high-frequency measurements in high-speed waveform measurements.

In the second session, the focus will shift from theory to practice. Presenters will share their practical experience from deployments at RTE France, the U.S. eastern interconnection, and cloud environments. Attendees will learn the practical considerations that have led to successful deployments and see examples of commercial tools in action.

By striking a balance between technical rigor and practical considerations, this tutorial will benefit anyone interested in bulk power system oscillation monitoring.