Openmod workshop Canada 2026

Workshop
, ,
3

Staging area 1

This staging area is intended for potential contributors. Please feel free to edit and update the description of your intended in person contribution as your ideas evolve. Some of the entries listed here may still be at an early stage of development. Not all suggestions will necessarily be selected or developed into lightning talks or presentations.

The deadline for submitting your potential contribution is March 31st 2026. After this deadline passes, the organizing group will screen the suggestions and develop the final program.

This is a wikipost that anyone registered with the forum can edit. You are encouraged to add your potential contributions directly here therefore. The order given is chronological downwards.

*Note also that the topics and postings on this forum related to this event will be reorganized as the content builds. So just be aware that some URLs will break going forward.

Lightning talks:

Each lightening talk consists of a 6 minutes presentation followed by 4 minutes of Q&A. Profile your favorite project, tool, data, research findings, etc

When adding your lightening talk contribution, it should look something similiar to the contributions added below

Proposed contributions (please add your talk below :down_arrow: following the same template/style as the ones already posted)

:one: title : Coupling energy system models with life-cycle assessment - a case study on Québec energy system
presenter : Matthieu Souttre (CIRAIG, Polytechnique Montréal)
description : Energy System Models (ESMs) optimize transition trajectories but typically overlook environmental trade-offs beyond CO₂ — such as resource use, toxicity, or ecosystem impacts. This presentation introduces mescal, a model-agnostic open-source Python package that bridges ESMs with Life-Cycle Assessment (LCA) to embed these broader sustainability metrics directly into energy planning. As a case study, we present its application to the Québec energy system using EnergyScope, illustrating how prospective and regionalized LCA metrics can reshape transition pathways.
code repository : mescal (GitHub - matthieu-str/mescal: Coupling Energy System Models with Life Cycle Assessment · GitHub), EnergyScope (EnergyScope / EnergyScope · GitLab)
documentation : mescal (In a nutshell — mescal), EnergyScope (EnergyScope - Energyscope)
license : mescal - MIT, EnergyScope - Apache 2.0
literature :

:two: title : Advancing Renewable Integration through Demand Flexibility and Sector Coupling: a Systematic approach to unlock renewable energy’s full potential
presenter : Alebachew Mossie
description : Large-scale Renewable Energy (RE) integration faces significant challenges, particularly variability, intermittency, and grid stability issues which limit its full deployment. This presentation introduces a systematic approach to enhancing RE integration by leveraging demand flexibility through sector coupling. The concept focuses on linking electricity with flexible demand sectors, such as electric mobility and green hydrogen production, to better balance supply and demand, aiming to reduce RE curtailment, improve system efficiency, and enhance resilience. The work is based on an ongoing research proposal and outlines the proposed modelling framework, key research questions, and expected contributions to clean energy transitions, particularly in emerging energy systems.
code repository : work in progress
literature : flexible energy demands, Energy modeling tools, renewable energy

:three: title: Beyond the numbers: Examining economic modeling tools for climate change policy analysis
presenter: Madanmohan Ghosh
description: Over the past few decades, several economic models have emerged to assess the impact of policies addressing climate change. These models fall into five main categories: the integrated energy system model (IESM), the computable general equilibrium (CGE) model, the macroeconomic model (MEM), the agent-based model (ABM), and the integrated assessment model (IAM). This presentation provides an overview of these models, discusses key insights, and explores how they complement each other in climate policy assessments.
code repository:
literature: Energy-emissions-economy modeling, climate change

:four: title: Energy system implications of decarbonized truck electricity demand profiles
presenter: Lih Wei Yeow
description: The transition to low-GHG emission trucking and wider energy system decarbonization will reshape temporal energy demand and interact with variable wind and solar PV generation. Yet, comparisons between these truck decarbonization technologies seldom incorporate cross-sectoral interactions with the broader energy system at high temporal (e.g. hourly) resolution. Using year-long GPS data of 8000+ trucks in Ontario, Canada, and the Temoa open-source energy system model, we investigate how high-resolution, hourly energy demand profiles and cross-sectoral interactions influence cost-optimal technology deployment in a decarbonized energy system.
code repository: work in progress
literature: temporal demand profiles, heavy-duty trucks

:five: title: Rethinking European Energy Planning: What Open-Source Models Reveal About Network Development and Flexibility Needs
presenter: Luciana Marques (Open Energy Transition)
description: Open-source energy models are moving from research into real planning contexts, challenging the dominance of established commercial tools. In Europe, where transparency and traceability are becoming central to energy decision-making, this shift raises a key question: can open models deliver results robust enough for network development planning and flexibility needs assessment—and be trusted by stakeholders such as TSOs and regulators? Focusing on applications with PyPSA, this talk highlights how open-source approaches enable fully transparent and reproducible planning workflows. Drawing on validation against real-world open data, such as the Pan-European Market Modelling Database, it explores both the progress achieved and the remaining gaps, providing a concrete perspective on the role of open models in future European planning practice.
code repository: OpenTYNDP, FlexStudy
documentation: OpenTYNDP
license: MIT
literature: Network Development Plan (NDP), Flexibility Needs Assessment (FNA)

:six: title: Open Grid Data Matters: PyPSA Lessons from Colombia Using MapYourGrid data.
presenter: David Díaz (Open Energy Transition)
description: Open grid data quality can strongly affect energy system planning results. This lightning talk presents a PyPSA-based Colombia case studyusing transmission data improved through the MapYourGrid project.The study compares capacity expansion outcomes across three versions of the grid dataset (OSM 2020, OSM 2024, and latest OSM) under 2030 and 2050 scenarios with changing assumptions. The talk highlights the planning pitfalls that arise when grid data is incomplete, including distorted bottlenecks, misleading investment signals, and different expansion pathways.The session will also briefly show how the model was validated against external reference data and reflect on why collaborative open mapping should be treated as a core input to credible open-source power system modelling.
code repository: GitHub - pypsa-meets-earth/pypsa-earth-osm: PyPSA-Earth-OSM: Exploring Synergies between OSM data and energy planning · GitHub
documentation: Colombia Grid Impact Study
license: MIT

:seven: title: Uncertainty Management in Capacity Expansion and Resource Adequacy Studies
presenter: Dheepak Krishnamurthy (EPRI Canada)
description: Resource planners often need a model that is simpler and faster than a full production-cost or market simulation, but still rich enough to test capacity expansion decisions against resource adequacy outcomes. The talk focuses on a sandbox model for capacity expansion planning and resource adequacy studies, where this kind of model is most useful: rapid scenario screening, stress-testing planning assumptions, and connecting long-term portfolio choices to reliability outcomes.
code repository:
literature: Stochastic modelling, uncertainty management, energy modeling tools

:eight: title : Enabling multi- and many-objective energy system optimization with Osier.
presenter : Samuel Dotson (University of Illinois)
description : Conventional energy system optimization models (ESOMs) optimize a single objective function that typically represents an aggregated cost metric. However, stakeholders have and express preferences over many dimensions simultaneously and decisions are rarely made on the basis of cost alone. Osier fills this gap by offering three critical advancements.

  1. Osier can optimize an arbitrary number of user-defined objectives.
  2. An n-dimensional extension for modeling-to-generate-alternatives.
  3. Two dispatch algorithm formulations (logic-based and linear program).

Together, these advancements allow users of Osier to address structural and the lesser known, normative uncertainties present in all modeling exercises. These features are demonstrated on an example system and discussed.
code repository : Osier (github - arfc/osier: Justice oriented energy system optimization framework)
documentation : Osier (osier — osier 0.4.1 documentation)
license : BSD-3
literature :

  • Dotson, Samuel G., and Madicken Munk. 2024. “Osier: A Python Package for Multi-Objective Energy System Optimization.” Journal of Open Source Software 9 (104): 6919. https://doi.org/10.21105/joss.06919.
  • Dotson, Samuel Gant, Madicken Munk, and Kathryn Dorsey Huff. 2026. “Demonstrating the Osier Framework for Energy System and Nuclear Fuel Cycle Optimization.” Annals of Nuclear Energy 230 (June): 112151. https://doi.org/10.1016/j.anucene.2026.112151.

:nine: title : Simple models, real change: Using a PyPSA analysis to drive policy in Illinois.
presenter : Samuel Dotson (Union of Concerned Scientists)
description : Commonly, policy makers will propose specific targets for energy investment and subsequently direct relevant agencies to study the impacts of this proposed policy. This backwards approach leads to slower and less precise policy development. Instead, we find that energy system models should drive policy recommendations. This talk covers a case study in Illinois where a state-level energy system analysis by the Union of Concerned Scientists motivated specific policy guidance that was eventually adopted under the 2025 Clean and Reliable Grid Affordability Act (CRGA).
code repository : pypsa-illinois (github ucsusa/pypsa-illinois: A model of the Illinois electricity system built with PyPSA.)
license : BSD-3
literature :

  • Dotson, Samuel, Lee Shaver, and James Gignac. 2024. Storing the Future: A Modeling Analysis of Illinois Energy Storage Needs. Cambridge, MA: Union of Concerned Scientists. https://doi.org/10.7910/DVN/7QRME4

:ten: title : PyPSA-Canada framework and national model (Lightning Talk and Technical Demonstration)
presenter : Nathan de Matos, Michel Bui, Adrien Prigent (Natural Resources Canada, CanmetENERGY)
description : The lightning talk will introduce the soon-to-be-released PyPSA-Canada framework and national electricity model. This is a model of Canada’s interconnected electricity system, representing major generation and supply centres and the transmission corridors connecting them. Together, the model and framework allow exploring strategies to decarbonize Canada’s electricity system, such as expanding renewables, strengthening interprovincial transmission, or electrifying transportation and heating. They also support identifying the most cost-effective pathways to meet future electricity demand given environmental, technical and policy constraints. The technical demonstration will offer a guided walkthrough of the model’s features and scenario example showcasing how the tool can be used for analysis and decision-support.
code repository : NRCan github (repo not yet available)

documentation: Available in future repo
license : MIT

:one::one: title : Assessing the Spatiotemporal Generalizability of Power Outage Prediction Models using GeoAI Foundation Models
presenter : Yamil Essus
description : Machine-learning based power outage prediction models are increasingly proposed as decision-support tools for disaster preparedness and response, yet their ability to generalize across regions and events remains poorly understood. We evaluate the spatiotemporal generalizability of power outage prediction models using publicly available data for the U.S. East Coast. Specifically, we assess model performance under multiple test selection strategies, including unfiltered random splits, leave-one-state-out, and leave-one-event-out designs, which increasingly approximate real-world deployment conditions. We conduct a series of experiments to test common assumptions in power outage prediction. We focus on three methodological decisions that we find have a large impact on generalizability of power outage prediction models. First, the strategy to select an out-of-sample test set from the available data, specifically regarding potential data leakage due to spatial and temporal correlation in covariates. Second, the selection of target variable, which is often an absolute measure of outage magnitude and is inherently linked to population size. Lastly, the use of static features that could potentially lead to the model not being applicable to new areas due to overfitting or memorization of location specific patterns. We also present evidence that inconsistency of chosen performance metrics, study areas and time periods makes it difficult to assess the performance of literature models as a whole, and the heavy use of proprietary and region specific data significantly hurts the reproducibility and generalizability of the results. Additionally, in order to address the high dimensionality of weather features, we applied transfer learning to build a machine learning model based on a recent Geospatial Artificial Intelligence (GeoAI) foundation model (Prithvi WxC) specifically built for generalizability of weather features.
code repository : TBD (geospatial processing + ML pipeline)
literature : Power Outage Prediction with ML, GeoAI, Generalizability of ML models

  • Kapoor S, Narayanan A, Leakage and the reproducibility crisis in machine-learning-based science, Patterns, 2023; 4