Photovoltaics is the conversion of light into electricity using semiconducting materials. A photovoltaic device is composed of several layers each with different material characteristics. Measurements need to be done on each layer separately and on the device as a whole to develop improved photovoltaic devices. This database including millions of optical, sheet resistance, composition, XRD, and Raman measurements on single layers, and over 800,000 current-voltage scans of photovoltaic devices which can accelerate photovoltaic research and development.
Accelerating photovoltaic R&D with machine learning
Machine learning algorithms in combination with novel experimental data can accelerate the process of materials design and discovery for photovoltaic applications. Using machine learning algorithms on experimental data we have discovered new materials for photovoltaic absorbers which double efficiency, got ‘materials recipes’ for synthesizing new materials, and built tools to automatically analyze new measurements to retrieve insights. This could help design new photovoltaic devices with better efficiencies and lifetime which will greatly affect the photovoltaic industry.
See what’s been done with machine learning for photovoltaics:
Li, Jiaming, et al. "Machine learning for solar irradiance forecasting of photovoltaic system." Renewable Energy 90 (2016): 542-553.