The Contribution of Data Science to the Energy Transition at Eneco

My first data science project in the team centered on delivering flexibility, a crucial aspect of electricity security in modern power systems. I developed a machine learning model using sensor data from a solar park to generate a reference signal, indicating the expected electricity production if the park were operating at full capacity. Due to the model, we can manage solar parks that currently cannot produce an accurate reference signal themselves or send it frequently enough. This means that we can increasingly deliver flexibility with renewable energy sources.

As a Data Scientist, I collaborate with hardware, software, and data engineers who maintain the MLOps, Cloud, and streaming platforms. I also work closely with VPP end-users, like energy traders and operators with strong quantitative backgrounds. My role involves diverse tasks, including data collection, cleaning, processing, model building, and evaluating results with stakeholders. We often start projects with one-day hackathons, quickly building and testing models to validate solutions. This way of working suits Eneco's can-do mentality: If you have a good idea, you quickly get the green light to execute it.

Over the past six months, I've had the opportunity to develop into an end-to-end data scientist at Eneco, involved in everything from exploration to implementation. I've learned to work with data warehouses like Snowflake, streaming platforms like Kafka, write clean code, containerize solutions, and use CI/CD pipelines to deliver complete software solutions. The best part? The vast amount of real-time data from Eneco's many assets. It's truly data heaven. My work directly impacts the energy system and contributes to Eneco's goal of being climate-neutral by 2035.