ABSTRACT
High voltage engineering relies heavily on experimental data. However, experiments are often costly and time consuming, and in this context, analysis of experimental data has always been of great importance to extract proper information. Because statistical analysis is independent of the phenomenon or objects under investigation, it is applicable to a variety of subjects to sort out their representative parameters and to derive models that help make predictions about their performance under different conditions. It is no surprise therefore, that statistical analysis has found its way into high voltage engineering as a means to plan testing and analyze experimental data (Hauschild and Mosch, 1992).
