PhD Dissertation Proposal Defense: Cooper Sigrist, The New Way Forward: A Learning-Augmented Approach to Sustainable and Cost-Effective Rooftop PV Deployment

Content

Abstract

Rooftop solar photovoltaic (PV) installations have grown rapidly around the world, thanks to falling prices and incentive policies that spur adoption. These installations play a critical role in reducing the CO2 emissions of electricity generation and thus are important for various climate goals. Using ZIP code-level data spanning the United States that details PV installations, energy gen- eration, electric grids, and demographics, we find that areas with the greatest potential for CO2 offset—often lower-income, coal-reliant regions—are underutilized in the current distribution of installations and will remain underutilized in near-term projections of rooftop PV adoption. Through a series of counterfactual “what-if” analyses, we show that changing current adoption trends could increase CO2 reduction by 40%, using multi-objective evolutionary learning, and that these benefits could remain stable in the face of competing priorities beyond the singular goal of CO2 reduction (e.g., distributional impacts). We further simulate the effects of different incentivization policies to determine a cost-efficient, plausible strategy for the federal government. Our analysis shows that incentivization up to 22.5% is cost-efficient for both energy generation and CO2 offsetting and that removing all incentivization would lead to dramatic decreases in both.

Advisor

Mohammad Hajiesmaili