Energy Systems Modeling

Teaching · REE 4301 / IE 5300 · UT Arlington · 2026–

Many of the most pressing energy challenges cross traditional disciplinary boundaries and call for systems models to analyze possible solutions. This course shows how to construct such models, with particular emphasis on models that combine engineering, economics, the natural sciences and policy, using optimization, simulation, machine learning, decision analysis, stochastic processes and dynamical systems. Students learn to build mathematical models of complex, real-world energy problems.

Interactive worked examples →
Interactive, one per module: a toy dispatch model, a day of demand with EVs, the merit order and LCOE, power flow on a three-bus Texas grid, storage and mineral supply chains, and one facility at two sizes.

Modules

Introduction: what a model is, its boundary, and a first linear program
Demand: residential, commercial, industrial and transportation
Generation: the electricity system, fossil fuels and other sources
Networks: transmission, distribution and pipelines
Storage and supply chains: batteries, oil and gas, other storage, and material flows
Facility decision: one site at two sizes: should it build its own generation?

Topics

Energy accounting, units, conversions and Sankey diagrams
Demand modeling, load profiles and representative days
Dispatch and investment economics: merit order, heat rate, LCOE, capital recovery
Network modeling: DC power flow, losses, distribution and pipelines
Storage and material supply chains
Reliability, risk and safety; engineering economy
Scenario and sensitivity analysis, and model validation

How it is taught and assessed

Four mini-projects, one per module (40%)
A final integrated project: a report and presentation that brings every module together (20%)
Two mini-exams (30%)
Assignments and participation (10%)

Tools. Python in Google Colab, and Excel. Generative AI is encouraged for coding, presentations and proofreading, but not for drafting reports or hand-worked homework.

Graduate sections

Graduate students take the same course at a higher standard, and must also:

Collect global energy and supply-chain data from international sources such as the IEA
Run research-grade models such as ReEDS, GenX, PyPSA and BEopt
Validate model results, and run scenario and sensitivity analysis
Produce publication-quality reports and presentations

Textbook

Introduction to Energy System Modeling, Erick Jones, Mavs Open Press in progress

A companion graduate course, Energy Systems and Critical Mineral Supply Chain Modeling (IE 6301), begins in Fall 2026.

Key concepts

Short definitions of the ideas the course is built on.

Model: A stylized, simplified mathematical representation of one slice of reality, built to answer a specific question.
Parameters and decision variables: Parameters are the data, fixed before the model is solved; decision variables are what the solver chooses.
Infeasible and unbounded: Infeasible: no answer satisfies every constraint. Unbounded: the objective can keep improving forever, which usually means a constraint is missing.
Capacity factor: A plant's actual output over a period divided by what it would produce running flat out for the whole period.
Heat rate: A thermal plant's efficiency: the fuel energy, in BTU, it burns to make one kWh of electricity. Lower is better.
Levelized cost of energy (LCOE): Capital, operating and fuel costs over a plant's life, discounted to today and divided by the energy it produces.
Merit order: Dispatching plants from cheapest to most expensive running cost until demand is met; the last plant needed sets the price.
Price taker and price maker: A facility too small to move the market price is a price taker; one large enough to move it is a price maker.

Interactive models

What a Grid Is Made Of: energy demand to generation to the metal and rock behind it
Why Supply Can't Catch Up: why recycling cannot ramp, and why a mine started today arrives in the 2040s
The Texas grid: where ERCOT's power is made and where demand is going
Accurate, and Still Wrong: why linking a planning model to a supply-chain model needs more than a good fit

Free tools and readings

Open and free resources used in or alongside the course.