Peter Wintoft

Calculation of geomagnetically induced currents in the 400 kV power grid in southern Sweden

Sweden has experienced many geomagnetically induced current (GIC) events in the past, which is obviously due to the high-latitude location of the country. The largest GIC, almost 300 A, was measured in southern Sweden in the earthing lead of a 400 kV transformer neutral during the magnetic storm on 6 April 2000. On 30 October 2003, the city of Malmo at the southern coast suffered from a power blackout caused by GIC, leaving 50,000 customers without electricity for about 20-50 min. We have developed a model that enables calculation of GIC in the southern Swedish 400 kV power grid. This work constitutes the first modeling effort of GIC in Sweden. The model is divided into two parts. The electric field is first derived using a ground conductivity model and geomagnetic recordings from nearby stations. The conductivity model is determined from a least squares fit between measured and calculated GIC. GIC are calculated using a power grid model consisting of the topology of the system and of the transformer, transmission line, and station earthing resistances as well as of the coordinates of the stations. To validate the model, we have compared measured and calculated GIC from one site. In total, 24 events in 1998 to 2000 were used. In general the agreement is satisfactory as the correct GIC order of magnitude is obtained by the model, which is usually enough for engineering applications.

Real time Kp predictions from solar wind data using neural networks

Abstract Multilayer feed-forward neural network models are developed to make three-hour predictions of the planetary magnetospheric Kp index. The input parameters for the networks are the Bz-component of the interplanetary magnetic field, the solar wind density n, and the solar wind velocity V, given as three-hour averages. The networks are trained with the error back-propagation algorithm on data sequences extracted from the 21st solar cycle. The result is a hybrid model consisting of two expert networks providing Kp predictions with an RMS error of 0.96 and a correlation of 0.76 in reference to the measured Kp values. This result can be compared with the linear correlation between V(t) and Kp(t + 3 hours) which is 0.47. The hybrid model is tested on geomagnetic storm events extracted from the 22nd solar cycle. The hybrid model is implemented and real time predictions of the planetary magnetospheric Kp index are available at http://www.astro.lu. se/-fredrikb.

Short-term prediction of fof2 using time-delay neural network

Abstract To test the ability and efficacy of neural networks in short-term prediction of ionospheric parameters, this study used the time series of the ionospheric foF2 data from Slough station during solar cycles 21 and 22. It describes different neural network architectures that led to similar conclusions on one-hour- ahead foF2 prediction. This prediction is compared with observations and results from linear and persistence models considered here as two special cases of the neural networks.

Ionospheric foF2 storm forecasting using neural networks

Abstract The ionosphere shows a large degree of variability on time scales from hours to the solar cycle length. This variation is associated with magnetospheric storms, the Earths rotation, the season, and the level of solar activity. To make accurate predictions of key ionospheric parameters all these variations must be considered. Neural networks, which are data driven non-linear models, are very useful for such tasks. In this work we examine if the F2 layer plasma frequency, foF2, at a single ionospheric station can be predicted 1 to 24 hours in advance by using information of past foF2 observations, magnetospheric activity, and time as inputs to neural networks. Particular attention has been paid to periods when great geomagnetic storms were in progress with the aim to develop a successful ionospheric storm forecasting tool.

A neural network study of the mapping from solar magnetic fields to the daily average solar wind velocity

Predictions of the daily solar wind velocity (V) at 1 AU from the flux tube expansion factor ƒs are examined with radial basis function neural networks. The flux tube expansion factor is calculated from the potential field model, using Wilcox Solar Observatory magnetograms, with the source surface placed at 2.5 solar radii. The time series extend over 20 years from 1976 to 1995 and consist of approximately 3000 daily values of ƒs and V. The correlation between monthly averages of 1/ƒs and V is 0.57, independent of the assumed Sun-Earth solar wind travel time τ. However, for daily averages the correlation drops to 0.38 with τ = 5 days. Even adjusting τ to match the observed velocity does not improve on the overall correlation. A time series of ƒs(t) extending over t − 4 to t is used as input to the neural network. The network is trained to predict the solar wind velocity V(t + 2) 2 days ahead. The overall correlation on a test set, not included in the training, is 0.53, and the root-mean-square error is 85 km/s. Although the increase is significant, the correlation is still low. However, by studying a number of test cases it is seen that high-speed streams originating from coronal holes are well predicted, while transient structures related to coronal mass ejections are not predicted. To go further, a more detailed description of the solar magnetic fields must be included. The potential field model does not describe the currents in the corona, and changes of the photospheric magnetic field from day to day are smoothed out. By examining the relative error of the calculated photospheric magnetic field and the observed field, it is shown that the correlation between 1/ƒs(t) and V(t + 5) increases to 0.47 for errors smaller than 20% and drops to 0.3 for errors larger than 34%.

O+ Escape During the Extreme Space Weather Event of 4–10 September 2017

We have investigated the consequences of extreme space weather on ion outflow from the polar ionosphere by analyzing the solar storm that occurred early September 2017, causing a severe geomagnetic ...