Exploring an Energy-Status-Data Set from Industrial Production

Hansen, Arved (2019) Exploring an Energy-Status-Data Set from Industrial Production (Bachelor thesis), Kiel University, Kiel, 66 pp

[img]
Preview
Text
bsc_arved_hansen_thesis.pdf

Download (2988Kb) | Preview

Abstract

In this thesis we explore an energy status data set from an industrial production environment. This is done to learn more about industrial energy consumption, which helps enterprises to reduce their costs and can assist in reducing the overall energy consumption for ecological reasons. The exploration contains a visual analysis of the data and a forecasting of the future energy consumption. In order to do this we use the Titan monitoring platform, which has a component to visualize energy consumption. This component delivers several possibilities for a visual analysis. To bring the data into the Titan platform we design a tool which in general processes an energy consumption data set but is built in a way that it functions as a part of the Titan platform. Furthermore, we use the infrastructure of the Titan platform for the forecasting. This way we can deploy the monitoring and the forecasting as one product. However, due to the microservice architecture of the Titan platform the monitoring and the forecasting can be deployed individually.

The processing of the data can be adjusted to different modes in order to improve visual analysis and live demonstrations of the Titan platform. In the visual analysis we discuss different patterns in the data. Furthermore, these patterns serve as a base for the forecasting as the patterns should be recognized by a forecasting. For the forecasting we implement two different prototypes, one based on a statistical model and the other one is using neural networks. However, only the statistical approach delivered results. This is due to the fact that the forecasting in this thesis is not a finished implementation but serves as a prototype for future implementations.

Document Type: Thesis (Bachelor thesis)
Thesis Advisors: Hasselbring, Wilhelm and Henning, Sören
Keywords: Power Consumption, Energy Data, Monitoring, Forecasting
Research affiliation: Kiel University > Software Engineering
Kiel University
Projects: TITAN
Date Deposited: 21 Oct 2019 12:42
Last Modified: 21 Oct 2019 12:49
URI: http://eprints.uni-kiel.de/id/eprint/48018

Actions (login required)

View Item View Item

Document Downloads

More statistics for this item...