Javascript must be enabled for the correct page display

Manage the complexity of electric networks

Wentink, Benjamin (2018) Manage the complexity of electric networks. Integration Project, Industrial Engineering and Management.

[img] Text
Bachelor_IEM_2018_BenjaminWentink.docx
Restricted to Registered users only

Download (1MB)

Abstract

Electric networks are experiencing the integration of a wide array of new electronic devices and renewable energies, resulting in varying large and too complex network models. In most cases, a complete model of the electric networks is rather complex, which are neither practical nor necessary for e.g. transient analysis, controller design, or simulation. The general solution to this kind of problem is applying a reduction order technique to establish a lower-dimensional model to approximate the behavior of the original complex one. Nowadays, several reduction order techniques are available to be used. However, these kinds of methods are never fully developed and always have room for improvement. Therefore, in this integration project, a reduced order model is going to be investigated to look for improvements. The method under investigation is the balanced truncation method, combined with large and sparse networks, using an RC ladder network as an input. The system will be implemented in MATLAB and dissected to get an overview of the different operations and to search for the bottleneck computations. These bottlenecks are replaced by the Penzl algorithm, relying on the ADI algorithm. Finally, several benchmark models will be released on the two different versions of the balanced truncation method with contrasting properties of the matrices.

Item Type: Thesis (Integration Project)
Supervisor:
Supervisor nameSupervisor E mail
Cheng, X.X.Cheng@rug.nl
Degree programme: Industrial Engineering and Management
Thesis type: Integration Project
Language: Dutch
Date Deposited: 15 Feb 2018
Last Modified: 20 Aug 2019 09:53
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/16472

Actions (login required)

View Item View Item