|
ARUM: Adaptive Production Management
Basic facts:
|
  |
Our achievements:
See the article in Lidove noviny, September 17, 2014 |
 |
Applying Performance Analysis Concepts for Event-Based Industrial Automation Systems
The growing complexity of today's industrial systems motivates many researchers to utilize Event-Based Industrial Automation Systems (EBIAS) that enable to flexibly integrate heterogeneous components. However, developers of EBIAS lack efficient methods to verify the non-functional reliability of such systems. This methodological gap complicates the design of complex commercial systems with strict performance requirements.
Our activities conducted in cooperation with Austrian Institute of Technology (AIT) described existing problems and proposed to tackle it with performance modeling. The adopted approach was illustrated on a use case dealing with integration of an application designed according to the IEC 61499 distributed automation standard interacting with a SCADA environment in an event-based manner.
In order to show how the performance models can help to identify problems in EBIAS a real-world scenario has been selected. The chosen application is an event-based automation system for the AIT Smart Grid laboratory SmartEST. This laboratory is a highly flexible and configurable 1 MW research and test/validation facility for Smart Grid systems. It allows to test power system and ICT components in a Smart Grid configuration but also the validation of the component's behavior.
Distributed Control System for Buildings
The cooperation of ENEA and CTU advances the concept of Smart Grids in the field of providing a flexible energy consumption management framework, which considers the specific needs, requirements, and potentials of energy consumers. The cooperation connects expertise of both institutions and leads to development of a distributed control system for buildings. The project adopts existing models of buildings (one building acts as one consumer) and uses them to create individual negotiation strategies. The goal is to develop a system, which will maximize the overall utility function. Informally saying, the solution should provide results that will be in agreement with common sense expectations.
Let’s use a simple example with two air-conditioned office buildings A and B. Building A has high quality insulation and automatic sunblind system, on the other hand the building B has poor insulation and no sun protection. In both buildings employees start to work at the same time. The energy distributor offers a better energy tariff if the peak current consumption always stays bellow certain level and this level would be exceeded if AC units of both buildings would be on at the same time. It means that the buildings have to coordinate the AC switching. Because building A has better parameters, it makes higher global pay off to pre-cool this one in advance instead of pre-cooling the building B, because the final effect is lower in the case of badly insulated building and it makes better profit to start the cooling when it is really needed.
Engineering of Dynamic Simulation Models for Complex Mechatronic Systems
Petr Novak is a member of the Christian Doppler Laboratory for Software Engineering Integration for Flexible Automation Systems (CDL-Flex) at the Vienna University of Technology. He is responsible for new algorithms for designing simulation models. The proposed solution is based on the extension of the well-proven bond-graph theory. Utilizing and evaluating use-cases both from the Czech Republic and Austria, the developed simulation modeling tools and algorithms are applicable on a large variety of industrial systems. This topic as well as related issues will be discussed in the PhD. thesis by Petr Novak, which he is currently finishing.
RA-DIC: Rockwell Automation Laboratory for Distributed Intelligent Control
RA-DIC laboratory deals with the trend of massive data production which is entering many domains including automation domain. The utilization of smart devices generates huge amount of data that need to be processed in real-time. The Big Data paradigm represents possible way how to capture, store and analyze this amount and kind of data. RA-DIC laboratory identified variability as the main problem which has to be solved. Semantic technologies are used to process variable data and achieve previously impossible results.
Main focus: