Geomagnetically Induced Currents and Electric Power System Operations: The Case of Finland during Solar Minimum

Forbes and St. Cyr (2008) have reported evidence of a statistical relationship between a proxy for geomagnetically induced currents (GICs) and various metrics of power grid operations in 12 power grids located in the mid-latitudes during the more active portion of the solar cycle. This paper examines the possible effects of GICs on power system operations in Finland during solar minimum. The analysis makes use of statistical methods that make it possible to disentangle the effects of GICs on power system operations from purely terrestrial factors.

Fingrid, the electricity transmission system operator in Finland, is responsible for balancing the supply and demand of electricity within Finland. In cooperation with other system operators it also facilitates the scheduling of transmission services between Finland and other electricity control areas. In discharging its responsibilities, Fingrid makes use of day-ahead forecasts of both generation and load.

The starting point of this paper is the observation that the errors in the day-ahead generation forecasts appear to be unusually large. Specifically, over the period 16 March 2008 through 31 December 2009, the root mean squared error in the generation forecasts was 5.6 percent of the mean level of generation while the root mean squared error in the day-ahead load forecast was a more respectable 3.8 percent. There are also large differences between Finland's scheduled and actual electricity flows with Sweden. These differences are known variously as unscheduled flows, loop flows, and inadvertent flows. These flows are ubiquitous and frequently large in absolute value on alternating current transmission systems. System operators view these flows as a significant operational issue given that the stability of an electricity control area requires that generation plus net electricity flows from other control areas equal load at all times. Over the period 2 February 2008 through 31 December 2009, the root mean squared differences between Finland's actual and scheduled exports to Sweden were equal to 122 percent of its mean level of actual exports to Sweden. Over this same period, the root mean squared differences between actual and scheduled imports from Sweden were approximately 69 percent of Finland's mean level of actual imports from Sweden.

This study addresses whether GICs contribute to these operational challenges. It is hypothesized that GIC-induced energy losses contribute to actual generation being less than forecasted. It is also hypothesized that GIC-induced energy losses and transmission constraints affect the conductivity of the power grid which in turn contributes to the differences between actual and scheduled electricity flows. The study uses local (groundbased) magnetometer data as a proxy for GICs to test these hypotheses taking into account the other factors that can affect these metrics. The results of this analysis are expected to be of interest to system operators, policy makers, and the general space weather community.