AI-Powered Predictive Model for Solar Power Investment Decision Making

Abstract

Residential Photovoltaic (PV) systems have been gaining increasing attention due to their potential to contribute significantly to sustainable energy solutions. This dissertation delves into an in-depth exploration of the economic viability of residential PV systems across diverse geographic locations in the United States, presenting a novel decision-support tool, the Residential Solar Calculator (RSC). The research is carried out through two sequential studies. The first study executes a comprehensive Cost-Benefit Analysis (CBA) across multiple geographic locations, revealing considerable variability in the financial returns of residential PV systems. Building on this foundation, the second study performs a rigorous statistical analysis of region-specific influential factors, leading to an optimized predictive model developed using a neural network algorithm and advanced machine learning techniques. This predictive model is instrumental in the creation of the RSC, a user-friendly tool providing homeowners with personalized and accurate investment advice for PV installations. The RSC successfully bridges the gap between research and practical application, promoting the wider adoption of residential solar installations. This research has made significant strides in enhancing the precision, reliability, and comprehensiveness of economic evaluations of residential PV systems, and holds promise for informing policy development, PV system design, and investment decision-making. Prospects for future research are robust and multifaceted, involving further refinements to the predictive model, expansions of the RSC's functionality, and exploration of other influencing factors on the economics of residential PV systems.