You are here
Project Name: Machine Learning for Solar Technology Portfolio Management
Funding Opportunity: Solar Energy Evolution and Diffusion Studies 2 – State Energy Strategies (SEEDS2-SES)
SunShot Subprogram: Soft Costs
Location: Arlington, VA
SunShot Award Amount: $699,940
Awardee Cost Share: N/A
This project creates a data-driven tool that describes the development of different solar technologies through the use of machine learning and text analytics. This tool identifies the types of variables, and the influencers that impact them, that enable a solar energy technology to transition across readiness levels using unique solar energy data sources. In doing so, the model helps explain prior technology transitions, as well as predicts the likelihood of a technology’s advancement to future readiness levels.
The project team is conducting an extensive literature review of readiness scales, including the Technology Readiness Level (TRL) indicator, which is the most common method of estimating technology maturity. The team is also interviewing stakeholders to create a scale tailored to solar energy development goals, moving toward a unified framework for photovoltaics (PV). This includes the development of a rubric for estimating development levels of a technology, eventually developing software for predicting transition.
This project will create a more consistent and standardized tool for capturing the expected returns from a set of different technologies in a portfolio, allowing for better explanation of the portfolio value in quantitative terms. By applying the TRL indicator, this project enables enhanced risk assessment in technology investments and a firmer basis for understanding how solar technology evolves over time.