Mathematics in the Repsol Universe
According to Pythagoras, "all is number." A statement that has always made sense, and never more so than today. Big Data and Data Science, virtual reality and augmented reality, the Internet of Things (IoT), bioIT... Maths is now behind everything, above all in the field of science and technology.
This is demonstrated by the expansion seen at the Advanced Mathematics Department of the Repsol Technology Lab. In barely a year it has grown from a group of six people to a total of 23 scientists focused on a very wide range of projects, most of them connected with efficiency and sustainability.
The fact is that innovative companies need mathematical science to identify and resolve problems, and define strategies. In order to address these needs, four Repsol mathematicians specializing in different fields are working on a number of projects which this scientific discipline has raised to another dimension. Development of predictive models, simulation projects, cognitive technology, energy management systems… The applications of mathematics are almost unlimited.
Five years ago, Repsol scientists set about using cognitive natural language processing technology to optimize oil production. This is a similar technique to the approach used by Google when it suggests a result before you have finished typing, based on previous searches or logical sequences of words.
This reflection gave rise to the Pegasus Project, which has for the first time combined Repsol's crude oil production technology with the cognitive technology developed by IBM for natural language processing. "We rely on advanced mathematical techniques to speed up the optimization process," explains Prashanth Nadukandi, a specialist in numerical methods applied to engineering.
The mathematician is convinced that "within five or six years it will be possible to apply the technology to other fields, such as energy management." What most appeals to him about pursuing his career at Repsol is that his work "has an immediate or short-term industrial application."
Vasyl Gnyedykh specializes in the mass processing of information drawn from physical phenomena and industrial observations. His work centers on "energy efficiency, or resource management optimization," he explains, although there are many fields of application.
Data can be used to create prediction and simulation models. The big attraction for Vasyl is that "new IT techniques allow a huge volume of information to be processed quickly and efficiently en masse, in a way that would otherwise be difficult to handle."
Elena Núñez is working on a logistics project with the aim of minimizing transport costs. This requires a "mathematical definition of the processes and the development of an algorithm to help in decision-making."
The advantage of the system is "the possibility of incorporating uncertainty." These mathematical models are used to consider potential scenarios, which are resolved by means of the best solution for all the scenarios as a whole.
Ángel Rivero works on projects to simulate physical systems (an oil well, a chemical reactor or a car battery). How does he do it? "We aim to use equations to reproduce each of the mechanisms at work in the system. Such as waves, diffusion, collective transport, interactions among molecules or fluid phases, phase transitions, etc.," he explains. The appeal of these simulations is that by analyzing the solution on the computer, "we can predict future behaviors or draw design guidelines," Ángel explains.
Data Science helps us to be more efficient, simulation is used to predict future situations, algorithms are created to save costs, and the development of predictive languages helps us, for example, to care for the environment. The work undertaken by Ángel, Elena, Vasyl and Prashanth is just one example of how mathematics are increasingly needed at an innovative energy company like Repsol.