An agent-based approach to designing residential renewable energy systems

Residential consumers in the U.S. have demonstrated a growing interest in rooftop photovoltaic (PV) systems, resulting in increased adoption over the last decade. However, this has diminished utility revenues, and policymakers have expressed concerns about inequitable consumer access to publicly-funded rooftop PV adoption incentives. In response to these concerns, policymakers and utility companies are changing their policies to discourage rooftop PV adoption. Alternative renewable energy models, such as utility-provided community solar programs, offer a potential solution. However, when designing such programs, it is important to consider the potential impacts on different system stakeholders, including utilities, policymakers, and solar installers. This paper describes an agent-based model that predicts the performance of different residential distributed solar models with respect to these stakeholders' objectives. In this model, consumer agents residing in an urban utility territory decide in each time-step whether they will adopt a particular renewable energy model, and the impacts of their adoption decisions on stakeholder performance metrics are captured over time. Simulation results suggest that if community solar program premium prices are set appropriately, all stakeholders can benefit: the utility can recover part of its revenue losses even as rooftop PV adoption increases, solar installers’ businesses can thrive, and increased renewable energy adoption can be achieved equitably. The proposed modeling methodology can help to inform design decisions of distributed solar energy models that avoid benefiting some stakeholders at the unnecessary expense of others.