Energie Campus Nürnberg - EnCN Simulation - Subproject 1: Hybrid Simulation of Intelligent Energy Systems
The Energie Campus Nürnberg (EnCN) researches the energy chain based on renewable energy.
It is a holistic approach, taking into consideration the entire energy chain, from generation over distribution to utilization.
The ongoing process of replacing fossil fuels and nuclear energy by renewable energy sources arises new challenges.
They include the examination of new materials, technologies, and processes for the production, storage,
and consumption of renewable energy.
They also include the challenge to overcome the top down approach of transporting and distributing energy,
currently tailored to particular needs of power generation with fossil fuels and nuclear energy.
The EnCN is a collaboration of the Friedrich-Alexander-University of Erlangen-Nuremberg,
the Georg Simon Ohm University of Applied Sciences Nuremberg, the Fraunhofer Institute for Integrated Circuits,
the Fraunhofer Institute for Integrated Systems and Device Technology,
and the Bavarian Center for Applied Energy Research in tandem with industry partners.
EnCN Simulation is one cross-sectional project of the EnCN, dealing with modeling, analysis, simulation,
and optimization in the energy chain context.
Other projects are EnCN Transport, EnCN Net, EnCN Process, EnCN Building, and EnCN Economy.
Within EnCN Simulation the sub project Hybrid Simulation of Intelligent Energy Systems
simulates smart grid scenarios to study the collaboration of
power plants (central and distributed, conventional and regenerative), consumers (conventional and adaptive),
and energy storages (pump water, rechargeable battery, chemical energy carrier, etc.).
We utilize discrete simulation for modeling the renewable energy sources
and continuous simulation to represent the dynamic effects of the smart grids.
The mass production of cars doesn't exclude the individualization of each car.
Mass-customization is nowadays important to meet the growing demand for customer satisfaction.
This results in an increasing complexity within the involved processes, information and data.
We have made use of multi-valued decision diagrams (MDDs) to ensure consistency within car manufacturing data.
The size of an MDD depends heavily on the variable order and thus the order of the layers in the MDD.
It was unfeasible to keep the possible product configurations encoded as an MDD in memory
without an appropriate ordering of the variables.
Even though with the application of well-known reordering techniques for the reduction of the memory consumption,
the memory size of interim MDDs during the construction process exceeded by far the available memory of the used PC.
Therefore we introduced a new reordering technique for the constraints.
In the domain of car manufacturing the constraints are used to express
that certain product configurations are not valid.
With the dynamic reordering of the constraints together with the dynamic reordering of the variables
the MDD, containing all possible product configurations, could be constructed.
The resulting MDD was used to check the consistency of other data, like the bill of materials,
and errors have been detected.
Analysis Methods for Non-Markovian Models
Traditional approaches to solve non-Markovian models use phase-type expansions,
apply the method of supplementary variables or construct an embedded Markov chain.
All three approaches have also been investigated in the context of queuing networks and stochastic Petri nets.
The phase-type expansion approach suffers from the enlargement of the state space,
whereas the method of supplementary variables and the embedded Markov chain construction require basically
that non-exponentially timed activities are not concurrent.
If they are concurrent this will result in multidimensional differential equations that are hard to solve.
To avoid these problems more efficient techniques for the performance evaluation of computer networks
like web servers or networks of embedded systems have to be developed.
In such systems activity durations with large variances (file transfers)
as well as deterministic durations (security aspects) arise.
We have created two new approaches to approximately evaluate the performance models of these systems;
the first one is based on the method of supplementary variables and the second one deals with phase-type expansions.
The method based on phase-type expansions is integrated in the tool WinPEPSY-QNS - a tool for performance evaluation
and prediction of queueing systems developed in cooperation with department 4 (Distributed Systems and Operating Systems).
WinPEPSY-QNS
(Windows Performance Evaluation and Prediction System for Queueing NetworkS)
The efficient modeling and performance evaluation of computer, communication, production,
or workflow-management systems needs powerful modeling methods, algorithms, and tools.
Queueing network models are very popular for modeling such systems and calculating performance measures.
Therefore the queueing network tool
WinPEPSY-QNS
with a modern GUI and facilities for
extensive experiments which can be run on Windows PCs has been developed.
It offers well known analysis methods as well as newly developed methods
for product-form and non-product-form queueing networks.
Dipl.-Inf. Peter Bazan
