Solar activity and transformer failures in the Greek national electric grid

Abstract

We study both the short term and long term effects of solar activity on the large transformers (150kV and 400kV) of the Greek national electric grid. We use data analysis and various analytic and statistical methods and models. Contrary to the common belief in PPC Greece, we see that there are considerable both short term (immediate) and long term effects of solar activity onto large transformers in a mid-latitude country (latitude approx. 35 - 41 degrees North) like Greece. Our results can be summarized as follows: For the short term effects: During 1989-2010 there were 43 stormy days (namely days with for example Ap larger or equal to 100) and we had 19 failures occurring during a stormy day plus or minus 3 days and 51 failures occurring during a stormy day plus or minus 7 days. All these failures can be directly related to Geomagnetically Induced Currents (GICs). Explicit cases are presented. For the long term effects we have two main results: The annual transformer failure number for the period of study 1989-2010 follows the solar activity pattern (11 year periodicity, bell-shaped graph). Yet the maximum number of transformer failures occur 3-4 years after the maximum of solar activity. There is statistical correlation between solar activity expressed using various newly defined long term solar activity indices and the annual number of transformer failures. These new long term solar activity indices were defined using both local (from geomagnetic stations in Greece) and global (planetary averages) geomagnetic data. Applying both linear and non-linear statistical regression we compute the regression equations and the corresponding coefficients of determination.