Repsol teams with Google Cloud to optimize refinery using big data, artificial intelligence

The project has the potential to add 30 cents on the dollar to Repsol’s refined barrel margin, which could translate to 20 million dollars annually for the Tarragona refinery, with significant upward growth if all optimization objectives are achieved, Repsol said.

The initiative puts the latest cloud technology from Google at the service of the refinery’s operators. Repsol’s objectives are to maximize efficiency, both in energy consumption as well as consumption of other resources, and to improve performance of the refinery’s overall operations.

The project marks a pioneering challenge in the global refining industry, the company said, noting refineries are among the largest and most complex industrial facilities. Tarragona is one of the six refineries that Repsol operates in Spain and Peru. Repsol’s third-largest unit, the plant has capacity to distill 186,000 barrels of oil a day. The refining unit processes 9.5 million tonnes of raw material a year and the storage tanks can hold a million cubic metres.

Google will make available to Repsol its data and analytics products, the experience of its professional services consultants and its machine learning managed service, Google Cloud ML, which will help Repsol’s developers to build and bring machine learning models to production in their refinery environment.

The management of a refinery involves around 400 variables, which demands a high level of computational capacity and a vast amount of data control. This is an unprecedented challenge in the refining world, Repsol said.

10 times variables being managed 

Until now, the highest number of functions integrated digitally in an industrial plant is around 30 variables, demonstrating the vast challenge the project presents. It aims to increase the number of variables being managed by more than 10 times.

Repsol said it chose the Tarragona refinery to develop this initiative because the online configuration of its production schematics facilitates testing and implementation.

The project, as well as the collaboration with Google Cloud, is part of Repsol’s ongoing digitalization, innovation and technology projects development in all of its business units to improve its competitiveness and efficiency.

“This is an efficiency project in all senses: it seeks to consume less resources; reduce energy consumption, which is the highest cost of a refinery; increase the unit reliability and, by extension, improve economic performance," Maria Victoria Zingoni, Repsol’s executive managing director of Downstream. said in a statement.

“This initiative belongs to a more comprehensive plan to take advantage of the possibilities afforded us by the latest in technology, and improve industrial processes. We are not afraid of aiming for the stars, even if some projects will fail. This is about learning as fast as possible and that machines help people in their work."

"This project demonstrates the commitment from Spanish companies to digital transformation and the application of machine learning in industrial processes, of which Repsol is a pioneer,” added Fuencisla Clemares, country manager Google España y Portugal. “At Google we are deeply committed to sustainability and ensuring that we have a positive impact on the environment – and we see technology such as machine learning and data analytics playing an important role in helping our customers maximize their own efficiency.”

Source: http://www.jwnenergy.com/article/2018/6/repsol-teams-google-cloud-optimize-refinery-using-big-data-artificial-intelligence/