Input requested about Data Taxonomy to Support PV Performance Modeling

Greetings,

Sandia National Laboratories is proposing a project to create a data taxonomy to support PV performance modeling, and we’d welcome your thoughts.  The proposed project is briefly described below. Many of you have PV performance modeling infrastructure in place and we’re particularly interested to know how our proposed work could add value to your work.

Questions:  Please email responses to Dr. Cliff Hansen (cwhanse@sandia.gov)

1. If the taxonomy and accompanying software tools were available today, would your organization use it?
2. Which of these applications of the proposed taxonomy do you see as having the greatest value?
   1. Managing sets of PV model parameters
   2. Archive and retrieval of historical weather and/or system data
   3. Archive and retrieval of module and/or system data
3. Briefly indicate how your organization currently manages data for PV performance modeling.
   1. We have databases to contain model and weather data.
   2. We have a databases or a file system (e.g., directory structure by project or location) to manage performance modeling files.
   3. We don’t have a formal system to manage performance modeling data.
   4. We manage performance modeling data using _____________________________
   5. We don’t do PV performance modeling.
4. Any other comments you’d like to offer regarding the proposed project:

Project Description

A data taxonomy for PV performance modeling would provide a consistent, well documented structure to store, retrieve and exchange the following:

• Parameters for performance models (e.g., module model coefficients, inverter model coefficients)
• Historical weather and system measurements
• Data from laboratory or field characterization of PV components or systems (e.g., IV curves)

We hope that a formal taxonomy for these data will facilitate transparency and reproducibility in performance modeling, as well as reduce time and costs associated with exchange of modeling information.  Development would proceed through public working groups, and the resulting taxonomy would be a public release with implementation for common languages such as python and java, as open source code.  We could also consider writing utility code to translate between the taxonomy structure and various popular formats for data, such as TMY3 for weather data, PVsyst’s PAN files, and the SAM libraries of CEC model coefficients.

Thank you,

Cliff Hansen

Sandia National Laboratories

About the author

Joshua S. Stein is a Senior Scientist at Sandia National Laboratories.