The minds behind the algorithms

With machine learning algorithms, energy experts can show customers directly what savings they can achieve using our services.
  • Martin Eksberg

    Head of Business, Sweden

  • Filip Lagerlöf

    Marketing

published

updated

The minds behind the algorithm, Photo: Raghunath Vairamuth and Mitra Hajigholi
Photo: Raghunath Vairamuth and Mitra Hajigholi, data scientists at Kiona.

Mitra Hajigholi and Raghunath Vairamuth are data scientists focusing on machine learning and data analysis with Kiona. They have worked on several projects connected to data science, such as Peak Control 2.0, AI-steering, and predictive maintenance.

What is Machine Learning?

Machine Learning (ML) is a type of artificial intelligence (AI) that allows the software to become better at predicting outcomes without being plainly programmed to do so. Almost like the way that humans learn, the accuracy will gradually improve. Read more about ML at Wikipedia

One valuable result is our control algorithm, which allows us to see how energy consumption varies at different temperatures in real-time. Previously, energy consumption was analyzed and compared based on reference year data.

Customer insights

Mitra and Raghunath's work with the platform continues. One important aspect of this is to collect relevant data based on customers' needs for information.

Mitra has completed several interviews to determine what data customers want to see, are interested in, and why this information is vital to them. The data generated will be used to create new Machine Learning algorithms. This will allow us to develop better, intuitive, and visual reports that can be tailored for each customer.

We have also collected more weather-related data points to improve our control algorithm. In this way, we can help customers identify energy peaks on a larger scale.

Raghunath VairamuthData Scientist, Kiona

We have also collected more weather-related data points to improve our control algorithm. In this way, we can help customers identify energy peaks on a larger scale. We can also proactively warn energy companies days in advance, says Raghunath.

Using data to increase energy efficiency

In addition to the above, Mitra and Raghunath are working on improving predictive maintenance. Here, the data is used to analyze the central heating performance. Advanced analysis of the amount of heat that enters and leaves the heating system makes it possible to detect leaks or deviations.

– We need to help customers understand the value of collecting correct data from their building portfolio and what type of data is valuable. That is when it becomes a win-win-win from an environmental, customer, and cost perspective, Mitra concludes.

Want to know more about our self-learning AI engine?