Transmission Lines in Poland and Space Weather Effects

We analyze the solar influence of space weather events on Polish energy infrastructure via geomagnetic disturbances. We consider two time intervals of very different levels of solar activity (SA) in the course of the solar activity cycle 24: 2010—an early rising phase of SA, near the solar minimum and 2014—solar maximum. From the total number of electrical grid failures, we extract almost five thousand failures in 2010 and more than ten thousand in 2014, which could be related to space weather effects. We analyze data of failures having unidentified causes, failures coupled with the aging of transmission network elements and electronic devices, which appeared during the time intervals of enhanced geomagnetic activity. We show that around the times of increasing geomagnetic activity level, the amount of transmission lines failures in the groups mentioned above grows, but mostly with some delay. It might be connected to some cumulative effect due to the result of transient states and their propagation in the distribution network. Although we do not exclude that the other factors may play a role, the solar-originating disturbances strengthen these effects.


Introduction
The capacity installed in the Polish National Power System in 2017 was 43,332 MW increasing by 5% in relation to 2016. According to the Energy Regulatory Office, the volume of gross domestic electricity production in 2017 was at a higher level than during the previous year being 165,852 GWh, which is an increase of 1.98% [1]. This tendency is kept-Polish industry, as well as Polish society electricity demands, are continuously growing [2]. Although, due to redundancy, the blackouts in the Polish system are infrequent events, each factor, which possibly may increase the probability of their occurrence, should be considered [3].
One of these factors is solar-driven effects, which can continuously influence the near-Earth's environment, hence the space-born and terrestrial technological systems, for example, [4], and as a result, humankind existence, for example, [5,6], as well. Sun is changing its behavior in various time scales. A sunspot number generally categorizes solar activity (SA). Sunspots emerge on the surface locations [29]. The same storm was reported to generate serious damage of 15 transformers in South Africa [12,30].
The more recent event is the St. Patrick's Day geomagnetic storm on 17 March 2015, being the strongest in the 24th solar activity cycle [31]. Kozyreva et al. [32] reported that its main phase was rather step-wise (∼17 h duration), it contained shortly-living substorm-like activities, with geomagnetic indices, IE and AE, up to ∼1500-2500 nT (where AE index from http://wdc.kugi.kyoto-u.ac.jp, describes the planetary geomagnetic field disturbance at auroral latitudes, which are caused by the auroral electrojet and IE is the IMAGE Electrojet indicator from https://space.fmi.fi, being a simple estimator of the total eastward and westward currents crossing the magnetometer network). This event was initiated by the interplay between two CMEs and high-speed solar wind streams from a corotating interaction region [33] and caused the GIC enhancement in transmission lines on the Kola peninsula [32]. Though, the one-to-one coherence between the power of the substorm, described by the local electrojet index, and the power GIC was not found. Carter et al. [14] also examined the St. Patrick's storm and found that even at the around equator latitudes, strengthened GIC appearance was noticed, both at the moment of sudden storm commencement and about 10h after the storm, at similar moments as the greatest perturbations of the geomagnetic field in the high-latitudes areas. These authors explained the high-intensity GIC appearance around the equator by the magnetospheric, as well as ionospheric currents' fluctuations linked to a sudden growth of the solar wind dynamic pressure with interplanetary electric fields (IEF = −V × B z ) playing a minor role.
Over the last few decades, a number of authors have investigated the potential impact of severe space weather on energy infrastructure, for example, [34][35][36][37]. In particular, space weather phenomena may affect the physical infrastructure (e.g., transformers) required to handle electricity transmission and generate an energy collapse in the transmission lines, presumably announced by voltage stability loss, which will shield the power system's resources from damage. Schulte in den Bäumen et al. [38] showed that depletion of 10% in electricity supply in the Earth's immensely industrialized and most populous areas due to an extreme geomagnetic storm can impact the global economy in the same scale as global financial crises and wars. Schrijver et al. [39] found that insurance claim rates for industrial electrical equipment across North America greatly rose on days with strengthened geomagnetic activity.
The above-mentioned failures of energy facilities make society conscious that geomagnetic storms are the result of a very complex chain of events derived from the accumulated, in the interior of the Sun, magnetic energy [28]. In the process of the national power systems maintenance and development, it should be taken into account that geomagnetic storms are indeed genuine threats and can cause significant damage to both power systems and other critical infrastructure elements. If not appropriately handled, the socio-economic ramifications can be severe. Significant consequences for the power system are local or nationwide power outages, catastrophic transformer breakdowns, and long-term loss of production. Therefore, it is necessary to analyze whether and to what extent the space weather effects influence the Polish energy infrastructure by the enhanced GICs. This paper is organized as follows: in the introduction, we present a short overview of solar impact mirrored in geomagnetic disturbances. In the second section, we introduce the data used in this paper. The third section describes transmission lines in southern Poland, as well as reasons for their failures. In the fourth section, we present our results and discussion. The last section concludes the paper.

Geomagnetic Conditions in Poland in 2010 and 2014
Daily regular magnetic field variation arises from current systems caused by regular solar radiation changes. Other irregular current systems produce magnetic field changes caused by the interaction of the solar wind with the magnetosphere, by the magnetosphere itself, by the interactions between the magnetosphere and ionosphere, and by the ionosphere itself. Various indices were designed to describe variation in the geomagnetic field caused by these irregular current systems. Geomagnetic indices are a measure of geomagnetic activity, which is a signature of the response of the Earth magnetosphere and ionosphere to solar forcing. They play a significant role in describing the magnetic configuration of the Earth's ionized environment [40]. In the second half of the twentieth century, they have become a key parameter in solar-terrestrial studies; in the past 15 years, they have become a key parameter in space weather, being commonly used to detect and describe space weather events. Geomagnetic disturbances can be monitored by ground-based magnetic observatories recording the three magnetic field components. Magnetic observations are carried out by high precision magnetometers, for example, [41,42], that measure the total geomagnetic field intensity where B X , B Y and B Z are the Cartesian components of the geomagnetic field vector in the north X, east Y and down Z directions, respectively. Geomagnetic field measurements detected by magnetometers include the regular and irregular (contributing to K index) time variations of geomagnetic activity.
One of the essential indexes of global geomagnetic activity is the planetary Kp index. The planetary three-hour-range index Kp was introduced by J. Bartels in 1949 and is derived from the standardized local K index of 13 magnetic observatories between 44 degrees and 60 degrees northern or southern geomagnetic latitude. It is designed to measure solar particle radiation by its magnetic effects, and today it is considered a proxy for the energy input from the solar wind to Earth. The local three-hour range K index is a number in the range of zero (quiet) to nine (disturbed) that classifies the observed variations of the geomagnetic field locally after subtraction of the regular daily variation. Each activity level relates almost logarithmically to the corresponding disturbance amplitude of the horizontal field component during a 3-h UT interval. In a day, eight K indices are given in successive 3-h UT intervals (0-3 h, 3-6 h, ... , 21-24 h UT) [43].
According to NOAA Space Weather Prediction Center scales, the K index ≥5 may indicate geomagnetic storm appearance (National Oceanic and Atmospheric Administration, NOAA, scale: https://www.swpc.noaa.gov/noaa-scales-explanation). This scale distinguishes various possible effects depending on the K index value. Thus, for K = 5, we can expect weak power grid fluctuations; for K = 6, high-latitude power systems may experience voltage alarms, long-duration storms may cause transformer damage; for K = 7, voltage corrections may be required, false alarms triggered on some protection devices; for K = 8, possible widespread voltage control problems and some protective systems will mistakenly trip out key assets from the grid might occur; while for K = 9, widespread voltage control problems and protective system problems can occur, some grid systems may experience complete collapse or blackouts. Transformers may experience damage.
To study precisely geomagnetic conditions in Poland in 2010 and January-July 2014, we have analyzed data from Belsk observatory [41], which is a part the of International Real-time Magnetic Observatory Network (INTERMAGNET). The INTERMAGNET is a consortium of geophysical institutes around the world connecting the action of geomagnetic observatories, for example, [44], monitoring different physical processes in the Earth's interior and the Earth's environment, for example, [42]. Nowadays, the INTERMAGNET consortium incorporates 57 institutes from 40 countries supporting 120 observatories, for example, [45] (Figure 1a,b therein). It provides the homogeneous observations of the terrestrial effects of solar corpuscular radiation by measuring the intensity of the geomagnetic activity caused by the electric currents produced around the Earth by that radiation [43]. Physically, the magnetic disturbance is primarily attributed to electric currents flowing in the ionosphere. Electric currents generate magnetic fields that contribute to the geomagnetic field measured by an observatory on Earth. For more amounts of events, it is sufficient to consider only three circular current-circuits [43]: the equatorial ring-current responsible for most of the world-wide effects, and currents in the two polar auroral zones, normally with about 23 degrees radius around the geomagnetic poles, but increasing in radius during intense storms.
We took into consideration geomagnetic index K from Belsk [46] in 2010 and 2014. Since K ≥ 5 may indicate geomagnetic storm appearance, we recognized all of the three-hour intervals with K index from Belsk being greater than or equal to 5. There were 31 events in 2010 and 26 during January-July of 2014 (from the entire number of 53 in the whole of 2014). An event with the highest K index value, being equal to 7, during the studied period was on 5 April 2010 (95th day of the year) at 9 a.m.

Characteristic Features of Transmission Lines in South Poland and General Reasons of Their Failures
Here we analyze failures of transmission lines in South Poland. This part of Poland is covered by tree type radial nets with a single power supply. There is also an option of backup power supply from another source point. It is accomplished by the Distribution System Operator (DSO) Tauron. DSO routines are based on modern technological solutions having the potential of guaranteeing the client's security of supply. To execute its responsibility, DSO advances and modernizes its energy infrastructure. Their transmission network consists of power lines, power stations, distribution transformers, branches, and street lighting. This electric energy system covers ∼25% of the electricity grids in the Polish National Power System [47]. The Polish National System of high voltage lines presents a map (Figure 1) arranged by the European Network of Transmission System Operators for Electricity (ENTSOE, www.entsoe.eu). ENTSOE represents forty-two electricity transmission system operators from thirty-five European countries.

Electrical Grids Failures' Classification
We have studied two-time intervals of the solar cycle 24: the whole year 2010 and January-July 2014. The first one was during an early rising phase of the solar activity cycle, around sunspot minimum and the second around solar maximum.
DSO logged 25,616 minor electrical grids failures (EGFs) in 2010. The failures' causes were divided into sixteen various categories (see details in [16]). Over the first seven months of 2014, the System Operator registered 30,155 minor failures. Based on data from the DSO reports (www.tauron.pl/ tauron/relacje-inwestorskie/raporty-okresowe in Polish), an increase appeared in the number of recipients of about 3.8% and the area covered by the distributor of about 7.4%. There was also growth noticed in the energy production of about 21.7% and in the length of transmission lines of about 25.7% between the years 2010 and 2014. Comparing the percentage of relative growth in the number of failures between 2010 and 2014, being around 50.4%, and the above-described development, we cannot expect that this increase in the number of failures was only due to the enlargement of the network.
The EGFs reasons were categorized into twenty-nine clusters. We have aggregated all these reasons into six, more general, groups from A to F. Similar categorization was used by Zois [18]. The EGFs occurrence rate in 2010 and in the first seven months of 2014 is presented in Table 1. The first three main groups (A-C) in Table 1 can be considered as impartial reasons. The next three (D-F), to some extent, can be linked to space weather effects. Details about solar activity impact may be found in [16,17]. Moreover, because meteorological effects (group A) in transmission lines failures were also more frequent in 2014, from meteorological observations and measurements provided by IMGW-PIB (https://dane.imgw.pl), we selected these meteorological effects, which were stated as the causes of failures in DSO reports. The data covered the southern part of Poland, and among them, there was information on the occurrence, duration, and strength of various types of effects. Detailed analyses have shown that in the first seven months of 2014, hail precipitation was 17 times more frequent than in the whole of 2010. Rime and snowfall appeared twice as often in the studied area in January-July 2014 than in the whole of 2010. During the first seven months of 2014, storms were 48% more frequent than in 2010, while gusty wind (15 m/s) appeared by 3% more often in 2014 than in 2010. The above data confirms the results contained in Table 1 that failures of power line components caused by meteorological effects were much more common in 2014 than in 2010.

Results and Discussion
Solar cycle 24 was much less active than, for instance, solar cycle 23, being somewhat analogous to cycles at the beginning of the XX century, when the Sun was pretty calm. In the course of this cycle, not many remarkable geomagnetic storms happened, and none of them was comparable to the Halloween Storm. During the examined time intervals, there were only a few cases when the geomagnetic K index from Belsk observatory was >5. The geomagnetic disturbances occurrence could not be indifferent to the functioning of the electrical infrastructure elements in Poland. Although there were no observed spectacular blackouts, the total number of electrical grids failures from general groups (D-F), which we suppose that might be connected, or at least strengthened by solar-induced phenomena (see Table 1), have grown around (with some delay) the time of geomagnetic disturbances. The mentioned delay (around one day) in EGFs emergence was shown by Gil et al. [16] using the method of superposed epoch analysis. This delay may be connected to some cumulative effect, for example, [18] due to the result of transient states and their propagation in the distribution network.
We compute geomagnetically induced currents based on the study by [48,49] with the induced surface geoelectric field according to the methodology introduced by Viljanen and Pirjola [50] and successfully used, for instance, to study St. Patrick's Day geomagnetic storm [51]. We had performed our computations for each case when the K index from Belsk was ≥5. For every event, we have considered 1-min data of geomagnetic field components B X and B Y from Belsk observatory at least one day before the first appearance of K value not less than five with an interesting event placed in the middle of the data-series.
We applied the uniform Earth's simple model with resistivity (ρ = 1/σ) information for the first ∼100 km layer [48]. By assuming that the earth is a half-space, we can use a Cartesian coordinate system with the x-axis directed to the north, y to the east and z downwards, for example, [50]. In a spatially uniform electric field E = E Y · e x + E Y · e y , GIC is given by formula GIC = a · E X + b · E Y , where e x , e y are unit vectors. Constants a, b depend only on the power grid parameters such as the geometry and resistances of the network [52]. Here we use (a, b) = (−62.3, 133.2) Akm/V after [53]. The horizontal electric field (E X , E Y ) on the surface of the earth along power lines is defined using the following equations [50]: where conductivity σ = 0.001(Ωm) −1 and magnetic permeability µ 0 = 12.57 × 10 −7 H/m, the electric field E Y is related to the magnetic field B X by function g X describing a derivative over time of the magnetic component B X (for simplicity of the equations we omit subscripts: B = B X and g = g X ) and with B n = B(T n ) denoting a mean value for the interval T n − ∆/2, . . . , T n + ∆/2 and ∆ = T n − T n−1 , (with n being an integer).
The same relation, of course, holds between E X and the geomagnetic field component B Y expressed by the function g Y . Thus, geoelectric field component can be written in the following way: with b n = B n − B n−1 and R N being an auxiliary function: We present, as examples, the computed values for the three longest events during the studied periods, i.e., 5 April (Figure 2), 2 May 2010 ( Figure 3) and 19 February 2014 (Figure 4). Analysis of all of the considered events showed that computed GICs grew after the time intervals when K index was greater than or equal to five (the second panel of Figures 2-4)       In Figure 2, the lowest panel presents the number of occurrence of electrical grids failures, which might have been caused by the space weather effects, at the beginning of April 2010. One can see that after the time when a high changeability in the geomagnetic field and thus in computed GIC appeared there was observed (with some delay) an unquestionable increase from the background null number of failures up to around 20 in the case of EGFs owing to the aging of transmission line elements and even 45 unidentified reasons. At the lowest panel of Figure 3, it is shown that there was a pulse rise up to 31 in the number of EGFs connected to the aging of infrastructure elements. At the lowest panel of Figure 4, it is depicted that during and after time intervals of strong geomagnetic fluctuations in February 2014, there was a strong growth of electrical grid failures' number in the categories D and F with the cumulative peak number equal to 162 in the group of failures due to the aging of the transmission line elements.

Conclusions
1. The assessment of the electrical grid failures' total number in the categories of failures, which might be connected to space weather effects near the solar minimum, i.e., in 2010 and around the sunspot maximum, January-July 2014, shows that the number of failures is twice larger in January-July 2014 than in 2010. Thus, it can be an indication of solar cycle phase dependency.