Simulation Precision and the Human Factor

Författare

  • Joni Klüss Aalto University Department of Electrical Engineering and Automation
  • Andrii Bichik Aalto University Department of Electrical Engineering and Automation
  • Ammar Arshad Aalto University Department of Electrical Engineering and Automation
  • Jan Strandberg Aalto University Department of Electrical Engineering and Automation
  • Syed Kazmi Aalto University Department of Electrical Engineering and Automation
  • Mika Laiho Aalto University Department of Electrical Engineering and Automation
  • Toumas Stark Aalto University Department of Electrical Engineering and Automation
  • Jay Panchal Aalto University Department of Electrical Engineering and Automation
  • Bilal Ahmad Aalto University Department of Electrical Engineering and Automation
  • Kenneth Väkeväinen Ensto Finland Oy

DOI:

https://doi.org/10.5324/nordis.v0i24.2287

Nyckelord:

FEM, field simulation, human factor

Abstract

Computer simulations are a powerful tool to support the design and development of electrical components and equipment. However, simulations are configured by a user, thus inevitably incorporating the human factor and potentially leading to divergence in results. To assess this degree of variance and thus the role of the user, 10 participants were asked to simulate the same medium voltage porcelain pin insulator using FEM software. Boundary conditions and materials were fixed. However, participants were able to define the geometry and details of the pin insulator by any means at their disposal. The varying skills of the participants resulted in geometries ranging from highly detailed complex models to rougher approximations. A CAD model provided by the pin insulator manufacturer is used as a reference. To quantify the extent of divergence, electric field intensity values in selected critical areas of the geometry are compared. This study presents the influence of the human factor and investigates the requirements for reliable simulation, i.e., how detailed does a model have to be to produce reliable information. The findings of this study can be used to save time and focus efforts on pertinent aspects in simulations.

Nedladdningar

Nedladdningsdata är inte tillgängliga än.

##submission.downloads##

Publicerad

2017-09-04