Case Study: TotalEnergies achieves faster global solar potential estimation with Google Cloud Platform

TotalEnergies Leading the charge in renewable energy through machine learning and data science

TotalEnergies, with support from Google Cloud Platform, wanted to speed up and scale solar potential estimation for its global service stations and other sites, especially in low-data regions where traditional tools were limited. Using Google Cloud Platform along with Google Earth Engine and Project Sunroof data, the company set out to build its Solar Mapper tool to better estimate solar output and support its decarbonization goals.

Google Cloud Platform helped TotalEnergies train and deploy machine learning models faster by switching to Cloud TPUs and using AI Platform and Cloud Storage. The result was a solar estimation tool developed in about six months, with test runs reduced from six days to six hours, and coverage in France expanded from 30% to 90%, enabling estimates in seconds instead of days.

TotalEnergies

Philippe Cordier

Research Program Director, Scientific Computing, Data Science and AI,