Long arc in still air: Testing, modeling, simulation and model parameter estimation

V. V. Terzija, H. J. Koglin

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


The paper presents the results of the investigation of a long arc in still air. In the first stage of the research, the long arc in still air is initiated under laboratory conditions in the high power test laboratory FGH-Mannheim (Germany). Based on the laboratory obtained arc voltage and the arc current data records, the main features of the arc are derived. In the second stage of the research the new arc model is developed. The arc is modeled as a current dependent voltage source with the characteristic waveform and it is considered as a source of higher harmonics distortion by the other signals in the network. Examples of computer simulations of an arc using new models are given. The main features of a simulated arc are described, too. Features of the real and modeled arc are given in the time and the spectral domains. The interaction between the arc and the network is analyzed, as well. In the third stage of the research, through the computer simulation typical examples of arcing faults on overhead lines are investigated. Finally, in the last stage of the research for the estimation of the unknown arc voltage model parameters the least error squares estimation method is used. The unknown model parameters are estimated from the computer simulated and laboratory obtained data.

Original languageEnglish
Article number896995
Pages (from-to)36-45
Number of pages10
JournalProceedings of International Conference on Harmonics and Quality of Power, ICHQP
Publication statusPublished - 2000
Externally publishedYes


  • Computational modeling
  • Computer simulation
  • Laboratories
  • Parameter estimation
  • Power system modeling
  • Power system simulation
  • Power system transients
  • Protective relaying
  • Testing
  • Voltage


Dive into the research topics of 'Long arc in still air: Testing, modeling, simulation and model parameter estimation'. Together they form a unique fingerprint.

Cite this