ORCID Identifier(s)

ORCID 0009-0005-4818-718X

Graduation Semester and Year

Fall 2025

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering

Department

Electrical Engineering

First Advisor

Ramtin Madani

Abstract

This dissertation develops interpretable, data-driven frameworks for short-term power demand forecasting using convex optimization and advanced feature engineering. The models combine historical load data, calendar structures, and meteorological variables to deliver accurate point, quantile, and probabilistic forecasts. By leveraging multi-periodic Fourier features, temperature-based regressors, and autoregressive memory, the proposed approach balances predictive performance with interpretability and scalability. Evaluations on multi-year datasets across US regions show consistent accuracy gains over benchmarks, while preserving transparency critical for real-world deployment. Beyond power systems, the framework generalizes to other time series applications in data science and AI, offering a robust, explainable alternative to black-box machine learning models for forecasting and optimization.

Keywords

Data-driven forecasting, Convex optimization, Interpretable machine learning, Probabilistic forecasting, Power demand prediction, Time series modeling, Feature engineering, Renewable energy integration, Artificial intelligence in energy systems, Scalable optimization

Disciplines

Electrical and Computer Engineering

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.