Correcting Electricity and Natural Gas Measurements using AI

Specialists at the metering companies make manual estimations of the energy usage over time, by looking at the energy volume, earlier usage of the client, and other factors. AI algorithm could possibly do the same thing. We want to learn more about Machine Learning algorithms and techniques, which may assist the specialists. We are especially curious about algorithms such as XGBoost and the family of RNNs.

Required interest(s)

  • Algorithms
  • Artificial Intelligence
  • Machine Learning
  • XHBoost
  • RNN's

What do you get

  • A challenging assignment within a practical environment
  • € 1000 compensation, € 500 + lease car or € 600 + living space
  • Professional guidance
  • Courses aimed at your graduation period
  • Support from our academic Research center at your disposal
  • Two vacation days per month

What you will do

  • 65% Research
  • 10% Analyze, design, realize
  • 25% Documentation

Large consumers of the Dutch energy grid, such as industrial clients, don’t get an energy meter from their regional grid operator. Instead, they need to choose a third party — a metering company — to conduct energy consumption measurements. These are used for billing and to give clients insight into their energy usage. For example, by measuring the electricity use every fifteen minutes and providing usage graphs and tables.

The metering devices installed by these companies usually allow the energy usage statistics to be accessed remotely. Most often this can be done without a hitch. However, sometimes the measurements seem to be missing or contain errors.

For billing this does not pose a problem, since a meter reading could still be done physically, to learn the consumed energy volume since the last reading. However, you lose insights in the exact energy usage over time (the 15 minute readings) by doing so. As a result, clients will lose valuable information on their energy consumption.

Currently, specialists at the metering companies make manual estimations of the energy usage over time, by looking at the energy volume, earlier usage of the client, and other factors. However, this requires large amounts of time from the specialists, while an AI algorithm could possibly do the same thing. Maybe even better than a trained engineer.

In this research we want to learn more about Machine Learning algorithms and techniques, which may assist the specialists in estimating the 15 minute readings, given the total energy usage of the time period for which the exact measurements are missing. Data on previous energy usage could be used as a training set to solve this task. We are especially curious about algorithms such as XGBoost and the family of RNNs. In addition, we would like to learn more about the (un)certainty of the recommendations made by the ML model, to give the specialists more clarity on whether a specific recommendation should be followed.

About Info Support Research Center

We anticipate on upcoming and future challenges and ensures our engineers develop cutting-edge solutions based on the latest scientific insights. Our research community proactively tackles emerging technologies. We do this in cooperation with renowned scientists, making sure that research teams are positioned and embedded throughout our organisation and our community, so that their insights are directly applied to our business. We truly believe in sharing knowledge, so we want to do this without any restrictions.

Sign up for this assignment

  • Geaccepteerde bestandstypen: docx, doc, txt, pdf.
  • Geaccepteerde bestandstypen: docx, doc, txt, pdf.

Application procedure

  1. 1
  2. Introductory meeting

    Discuss (study) career, interests and ambitions and introduction Info Support.

  1. 2
  2. Review

    Assessment of professional knowledge and personality (capacity, competences and motives).

  1. 3
  2. Selection interview

    Deepen professional knowledge and personality.

  1. 4
  2. The signing of a contract

    Contract offer and invitation for drawing moments.