ENERGY INFRASTRUCTURE SYSTEMS AS MULTI-SENSORS FOR PREDICTION AND DIAGNOSTICS(EASE)
![](/images/departments/32px/icon-h81000.png)
Abstract
The climate crisis is leading to an increase in extreme and unfavorable weather events and is having a major impact on the reliable operation of the energy infrastructure, be it through temporary events such as dark calms and icing or through damage caused by hail and storms. Both lead to high costs, on the one hand if balancing energy has to be provided unexpectedly, and on the other hand because insurance companies have to compensate for the damage. The protection standards that apply to infrastructure built today may no longer be sufficient in future climate scenarios and increased damage and, in the worst case, failures could occur. If these cannot be averted, they should at least be foreseen as precisely as possible in order to avoid worse things or to take timely measures. However, current methods do not make it possible to accurately predict local and short-term (extreme) weather events. Furthermore, information about the path and the time of arrival would be helpful for prediction and advance warning. On the one hand, to balance energy networks, and on the other hand, to prevent damage or to be able to react quickly. As part of the EASE project, a comprehensive system is to be researched in which historical and current data from geographically distributed energy infrastructure systems (e.g. PV, heat pumps, wind and water), weather forecasts and weather stations are networked with one another. This creates a dense 'sensor' network, which enables locally specific and precise short-term forecasts and an improvement in automated system diagnostics based on artificial intelligence (AI). Using AI, the huge amounts of data generated can be processed and potential benefits can be derived for various stakeholders. As a result, small-scale and short-term forecasts and diagnostics of generation plants are possible. From what happens in neighboring systems, it is learned by machine how parameters will develop locally at a target location, e.g. B. Cloud trajectories. On the one hand, the energy production of volatile renewables can be predicted with high accuracy and at the same time the development and path of (extreme) weather events can be recognized locally
Project staff
Philipp Weihs
Ao.Univ.Prof. Mag.rer.nat. Dr.rer.nat. Philipp Weihs
philipp.weihs@boku.ac.at
Tel: +43 1 47654-81424
BOKU Project Leader
01.11.2023 - 31.10.2026
Victoria Saval
Victoria Saval
victoria.saval@students.boku.ac.at
Project Staff
01.11.2023 - 31.10.2026
BOKU partners
![](/images/departments/60px/icon-h81000.png)
External partners
![](/images/partner.png)
AIT, Austrian Institute of Technology
Dr. Sarah Reisenbauer
coordinator