Identification of electrical parameters in a power network using genetic algorithms and transient measurements

• Wei Dong ^[1] ; Pericle Zanchetta ^[1] ; David W.P. Thomas ^[1]
  1. [1] University of Nottingham

     University of Nottingham

     Reino Unido

     
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 29, Nº 1 (Special Issue: Selected papers from EPE‐PEMC 2008), 2010, págs. 235-249
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – The knowledge of power network parameters (which are usually unknown or poorly quantified and cannot be measured directly) is very valuable for power system modelling, simulation, protection and control; since it is fundamental for solving many problems such as minimizing the effect of voltage distortion, active filter control or relay setting. The purpose of this paper is to develop a new, easy, effective and reliable method for power network parameters identification.

    Design/methodology/approach – The proposed identification method is based on current transient injections performed by a pulse width modulated power converter (which can be a rectifier or an active filter already installed in the network) and the analysis of the resultant network voltage response at the point of common coupling using genetic algorithms optimization.

    Findings – Simulation tests show the effectiveness of the proposed identification strategy achieving negligible errors, even in practical conditions when measurement noise is included. It is also a very flexible method since the optimization routine can be updated or modified in order to improve the performance.

    Originality/value – The developed method is non‐invasive, and uses only hardware already present on the power system, identifying the single parameters rather than the total impedance. It will provide improvements to several areas of power quality control. For example, when used within a stand‐alone piece of instrumentation, it will be possible to build up mesh equivalent networks of unknown power and distribution systems, which can then be used for load flow studies, protection or harmonic penetration prediction.