Powering global renewables with real-time data

THE CHALLENGE

Turning operational data into user-friendly information

Repsol Renewables is the global, low-carbon energy arm of Spanish oil and gas giant Repsol, with major operations in Spain, the US, Chile and Italy. It operates nearly 60 plants with over 300 wind turbines and 400 inverters, which convert a turbine’s power into grid-compatible electricity. And with each inverter alone generating around 1,000 data signals, the company handles huge volumes of raw data.

“My job is to add value to the immense amount of information we receive from each plant, each signal, and each piece of equipment, so it can be used by the company,” explains Juan Luis Vicuña, System Support Senior Consultant at Repsol Group.

A primary difficulty was allowing IT-friendly access to operational data. Specifically, Repsol needed to avoid a scenario where analysts had to use specialist operational technology (OT) tools to connect to equipment plant by plant, through isolated networks. Instead, it wanted a centralized system that would simplify data access and consumption.

“We also have a strategic roadmap to reach 7,000MW of installed capacity and potentially integrate new technologies like battery storage,” Vicuña added. “We needed a database that could scale efficiently to accommodate this growth without infrastructural bottlenecks.”

Storing every raw signal is costly, but Repsol must comply with regulatory requirements to retain information for set periods. A solution that offered efficient compression to reduce storage costs while maintaining data availability was, therefore, another key requirement.

“Different vendors, systems, and technologies force us to normalize models, so we don’t store data with inconsistent formats or semantic differences,” said Vicuña. “And in a global environment, visibility and security are also critical. Multiple network segmentations, firewall rules, and private links may increase security, but they also make daily work more complex.”

And with plants distributed across different continents, communication challenges and equipment failures also risked creating data gaps. Repsol required a cloud-based solution that ensured global connectivity and high availability through redundancy.

OUR SOLUTION

A simple, cost-effective solution with global connectivity

Repsol implemented MongoDB Atlas to bridge the gap between its OT and IT demands. And by moving away from an internal physical infrastructure to a managed, cloud-based environment, it also has the scalability it needs to ensure service continuity without manual hardware upgrades or adjustments to CPU power, RAM, and storage space.

“High availability through redundant nodes is very valuable,” said Vicuña. “And as there is no physical infrastructure, maintenance is simple and cost-effective. Deploying two clusters in Azure and AWS across different regions has been critical for integrating MongoDB Atlas with Repsol’s corporate architecture.”

Repsol uses MongoDB Time Series Collections for its five main datasets, two of which connect directly to high-frequency supervisory control and data acquisition (SCADA) systems. “These collections offer strong compression capabilities and faster analysis when properly indexed,” added Vicuña. “That drastically reduces our storage cost without compromising usability.”

This efficiency is complemented by MongoDB Online Archive, which allows Repsol to keep frequently accessed data on hand while retaining older data for regulatory compliance—and, critically, without bloating the database. “Two years of aggregated data in one collection only uses around 160GB,” added Vicuña. “And that storage volume remains stable thanks to MongoDB Online Archive, which also enables faster queries.”

The solution uses MongoDB Atlas Data API to serve data to internal and external clients, while an integration with Kafka as an ingestion layer prevents database overload and acts as a filter between the plants’ SCADA systems and MongoDB.

“We fine-tune ingestion because the initial loads caused issues due to excess data,” said Vicuña. “We now use Kafka to ensure that we access and store only useful data.”

OUTCOME

Centralized data simplifies consumption and analysis

The implementation of MongoDB Atlas has delivered significant operational, financial, and strategic improvements for Repsol. 

The business is successfully managing and leveraging the immense amount of information generated by its assets while keeping storage costs under control. In doing so, it is also making highly valuable data accessible to business users without requiring them to work through complex individual OT systems. And with Repsol’s Power BI instance accessing MongoDB data directly, users can then create their own reports independently.

“A renewables analyst doesn’t want to connect to each plant’s OT tools and navigate isolated networks,” said Vicuña. “By centralizing all that data, MongoDB Atlas is making consumption and analysis easy and effective.” 

Moving to the cloud with MongoDB Atlas has eliminated the need for physical infrastructure maintenance and ensures high availability, while autoscaling ensures continuity of service even during spikes in demand. 

The MongoDB Atlas deployment also provides the necessary global connectivity and redundancy to handle communications across plants in different countries, mitigating the risk of gaps caused by local failures.

The modernized architecture is now supporting the creation of future Repsol facilities, including a new control center in the United States.

“That requires the replication of our European cluster infrastructure and ensuring 24/7 real-time operations,” noted Vicuña. “We are also beginning to use MongoDB’s multi-region and multi-area capabilities. This global connectivity is proving to be a major advantage for Repsol.”