Using Machine Learning for Water Filtering Membranes Maintenance

Опубликовано 13 июля 2020, 16:23

The partnership between AWS and Veolia Water Technologies produced unprecedented results during an unprecedented time in our world’s history in eight short weeks. Anticipating when to clean or change water filtering membranes in desalination plants is complex.

Together, Veolia Water Technologies and AWS developed a prototype to optimize the timing of the maintenance. Historical time series data was fed into the Amazon DeepAR algorithm using Amazon SageMaker service, to learn from previous patterns and predict the future evolution of fouling indicators such as differential pressure or conductivity of the water.

This allows Veolia Water Technologies to reduce costs and prevent downtime while improving the quality of the water produced. The prototype is already in beta testing with plans to productize in the Middle East. It was developed during the COVID-19 pandemic, and the teams worked collaboratively from their home office.

These results couldn’t have been realized without the technical experience, absolute trust and dedication of both teams to achieve one goal: an uninterrupted clean and safe water supply.

Learn more about AWS Machine Learning at - amzn.to/38HpxXQ

Subscribe:
More AWS videos bit.ly/2O3zS75
More AWS events videos bit.ly/316g9t4

#AWS #AWSCustomer