The space project Sozvezdie-270 of Moscow University is in progress now. It involves the deployment of a CubeSat nanosatellites constellation. To the present, 20 satellites have been launched, 9 of them continue to function in near-Earth orbit; one more will be launched in the near future. Instruments were developed specifically for the experiments on board small spacecraft of the CubeSat format, which provide measurements of fluxes and spectrum of charged particles, primarily electrons of relativistic and sub-relativistic energies, as well as gamma quanta. Along with the space constellation, a network of ground receiving stations is also being created. A multi-satellite constellation gives a number of advantages in studying dynamic processes in near-Earth space. In particular, it makes it possible to carry out simultaneous measurements of charged particle fluxes with instruments of the same type at different points in near-Earth space. Such measurements provide unique information about the flux of sub-relativistic electrons, including variations due to precipitation of electrons, which is of great importance for understanding the mechanisms of acceleration and losses of trapped and quasi-trapped electrons in Earth’s radiation belts (ERB). We discuss various recent space weather manifestations associated with increased solar flare activity. Among such effects is the filling of the polar caps with particles of solar cosmic rays, dynamic processes in outer ERB during magnetic storms, rapid variations in electron fluxes due to precipitation. We discuss various recent space weather manifestations associated with increased solar flare activity. Among such effects is the filling of the polar caps with particles of solar cosmic rays, dynamic processes in outer ERB during magnetic storms, rapid variations in electron fluxes due to precipitation.

This study focuses on improving the neural network prediction of the geomagnetic indexes, in particular Dst-index, in a scenario, where input data is collected by two spacecraft (SC) with different data availability. One of the SC is approaching the end of its operational lifespan, while the other one lacks sufficient data history for constructing a high-quality neural network prediction. To effectively perform the transition between the two SC data, domain adaptation methods are needed. The study evaluates and compares various data translation techniques and optimizes the parameters for each translated feature to minimize domain discrepancies. The findings highlight the enhancement in the forecast, when employing domain adaptation methods and selecting relevant features, surpassing the results obtained using untranslated data.

Within the framework of the Moscow University space project SOZVEZDIE-270, a constellation of cubesat nano-satellites with a set of instruments is being deployed, which, among other goals, provides monitoring of the near-Earth space radiation environment, control of the geo- and heliophysical conditions. Along with the space constellation, a network of ground receiving stations is also being created. During the project implementation, 11 spacecraft of the cubesat format have been launched to date. Currently, there are 6 such spacecraft operating in near-Earth orbit, which transmit scientific and telemetric data. During 2023–2024 it is planned to launch at least 8 more such satellites into low circular polar orbits. Multi-satellite constellation has been implemented, which makes it possible to carry out simultaneous measurements of particle and quantum fluxes using the same type of instruments at different points in the near-Earth space. Such measurements provide unique information about the sub-relativistic electron flux dynamics, including variations due to precipitation, which is of great importance for understanding the mechanisms of trapped and quasi-trapped electron acceleration and losses.

In this paper, we study an algorithm for obtaining the most efficient model for predicting the amplitude of the geomagnetic Dst index, based on lowering the input data dimension by gradually discarding input features. This task is relevant, since the selection of significant input data is necessary for the effective use of machine learning methods. The study was carried out on the basis of the following machine learning methods: artificial neural network of the multilayer perceptron type, gradient boosting, linear regression. Comparison of the effectiveness of the listed methods is carried out.

One of the most effective tools for predicting time series is the use of machine learning methods, in particular, artificial neural networks. However, at the same time, a necessary stage of the study is to reduce the dimension of the input data. In this paper, we consider the results of data dimensionality reduction based on the ranking of input features by their significance in solving the problem of forecasting the geomagnetic Dst index. To assess the relative significance of features, an iterative approach is used, associated with the search of candidate models by discarding features one by one based on simple linear regression models.

Moscow State University is developing a project for a multi-satellite constellation intended for the monitoring of space radiation. A number of small satellites of CubeSat format were launched into selected orbits crossing the wide range of magnetic drift shells. The primary scope for the project is the operational monitoring of near-Earth’s radiation environment, i.e., fluxes of electrons and protons of Earth’s radiation belts and energetic particles of solar and galactic origin. To date, there are four CubeSat satellites operating in near-Earth orbits, which deliver scientific and telemetric data. Thus, for the first time, a unique multi-satellite constellation has been implemented, which makes it possible to simultaneously measure the particle and quantum fluxes at different areas in the near-Earth space using the same type of instruments. A special compact detector of gamma quanta and energetic charged particles (electrons and protons) DeCoR has been developed to carry out radiation monitoring by CubeSats. With their help, observations of various effects of space weather have been made. These effects include a variety of electron fluxes in the outer belt during geomagnetic activity in late November–early December 2021, filling of polar caps by solar energetic particles accelerated in flares occurring in late October–early November, and the existence of stable electron fluxes near the geomagnetic equator.

This paper performs comparative analysis of the prediction quality for the time series of hourly average flux of relativistic electrons (E > 2 meV) in the near-Earth space 1 to 24 h ahead by various machine learning methods. As inputs, all the predictive models used hourly average values of the parameters of solar wind and interplanetary magnetic field measured at the Lagrange point between the Sun and the Earth, the values of Dst and Kp geomagnetic indexes, and the values of the flux of relativistic electrons itself. The machine learning methods used for prediction were the multi-layer perceptron (MLP) type artificial neural networks, the decision tree (random forest) method, and gradient boosting. A comparison of the quality indicators of short-term forecasts with a horizon of one to 24 h showed that the best results were demonstrated by the MLP. The horizon of satisfactory forecast accuracy on independent data is 9 h, the horizon of acceptable accuracy is 12 h.

Internet-based system of Space Monitoring Data Center (SMDC) of Skobeltsyn Institute of Nuclear Physics of Moscow State University (SINP MSU) has been developed to predict and analyze radiation conditions in near-Earth space. This system contains satellite measurement databases and operational models and devoted to collect, store and process space weather monitoring data in the near real-time. SMDC operational services acquire data from ACE, SDO, GOES, Electro-L, Meteor-M satellites and use them for forecasting, now-casting and post-casting of space weather factors. This paper is intended to give overview of operational services of SMDC Internet-based system and demonstrate their possibilities and limitations to analyze space weather phenomena and predict radiation and geomagnetic conditions in the near-Earth space during February 14–March 5, 2014. This prolonged period of high level solar and geomagnetic activity demonstrates various manifestations of the space weather: solar proton events, geomagnetic storms and outer radiation belt (RB) dynamics. Solar sources of interplanetary space disturbances and their influence on geomagnetic and radiation state of the Earth’s magnetosphere were described using output coming from SMDC’ Web-based applications. Validation of SMDC’s operational models was performed based on the quality of description of the physical conditions in near-Earth space during space weather events observed from February 14 to March 5, 2014. The advantages and disadvantages of SMDC operational services are illustrated and discussed based on comparison with data obtained from satellites.

This study presents the results of prediction 1–3 days ahead for the daily maximum of hourly average values of relativistic electrons flux (E > 2 MeV) in the outer radiation belt of the Earth. The input physical variables were geomagnetic indexes, interplanetary magnetic field, solar wind velocity and proton density, special ultra-low frequency (ULF) indexes and hourly average values of relativistic electron flux. The phase-space for each physical component was reconstructed by time delay vectors with their own different embedding dimensions, and all of these vectors were concatenated. Next, various adaptive models were trained on this multivariate dataset. The following models were used for prediction: multi-dimensional autoregressive model, ensembles of decision trees within bagging approach, artificial neural networks of multi-layer perceptron type. The obtained results are analyzed and compared to the results of similar predictions by other authors. The best prediction quality was demonstrated by ensembles of decision trees. Also it has been demonstrated that using embedding depth based on autocorrelation function significantly improves prediction quality for one day prediction horizon.

Prediction of the time series of relativistic electrons fluence in the outer radiation belt of the Earth encounters problems caused by complexity and non-linearity of the “solar wind – the Earth’s magnetosphere” system. Artificial neural networks are a biologically inspired architecture that is a suitable tool to solve problems of such type. This study considers the dependence of the quality of prediction on the type and depth of delay embedding of input features.

GRB study is one of the main goals of the Lomonosov space mission. Multimessenger GRB observation is the only way of progress achievement in their study. The Lomonosov satellite is the first space mission in which the multi-wave length observations of GRBs are realized in real time without necessity of optical instrument re-orientation on GRB monitor trigger. The mission payload includes the GRB monitor BDRG, wide field optical cameras SHOK and UFFO instrument consisting of X-ray imaging telescope UBAT and UV slewing mirror telescope SMT. Two wide field cameras are placed in such a way that their fields of view (FOV) are overlapped by the GRB monitor detector FOVs. This allows the simultaneous GRB observations in gammas and optics in all-time scale of event evolution including obtaining optical light curves of prompt emission as well as of precursors. The real time data transfer to the GCN for detected GRB is realized as well as operative control of BDRG data on triggers from ground based facilities including neutrino and gravitation wave detectors. The GRB catalogue is presented in this report, also those events, for which information was obtained in different bands, are discussed in details.

AbstractSolar hard X-ray and gamma-ray emission was measured by BDRG instrument, the part of set of instruments operated on board the Russian satellite Lomonosov from April 2016 until now (solar-synchronous orbit with altitude 490 km, inclination of 97.6 degrees). Lomonosov measurements (11 flares with the X-ray energy more than 10 keV, and more than half of them have class in soft X-rays less than C2) were compared to the data obtained by RHESSI and Fermi space observatories as well as the Nobeyama Radioheliograph operating at the same time. The quasi-periodicity with different periods were found in some of them.

This paper presents analysis of the results of clusterization of the array of increases in the flux of relativistic electrons in the outer radiation belt of the Earth by two clustering algorithms. One of them is the algorithm for adaptive construction of hierarchical neural network classifiers developed by the authors, applied in clustering mode; the other one is the well-known k-means clusterization algorithm. The obtained clusters are analysed from the point of view of their possible matching to characteristic types of events, the partitions obtained by both methods are compared with each other.

Abstract Radiation environment of near-Earth space is one of the most important factors of space weather. Space Monitoring Data Center of Moscow State University provides operational monitor and forecast of radiation conditions both at Geostationary Orbits (GEO) and at Low Earths Orbits (LEO) of the near-Earth space using data of recent space missions (Vernov, CORONAS series) and current (Lomonosov, Meteor-M, Electro-L) ones. Internet portal of Space Monitoring Data Center of Skobeltsyn Institute of Nuclear Physics of Lomonosov Moscow State University (SINP MSU - [swx.sinp.msu.ru]) provides possibilities to monitor and analyze the space radiation conditions in the real time mode together with the geomagnetic and solar activity including hard X-ray and gamma-emission of solar flares.

The paper investigates the possibilities of the prediction of the time series of the flux of relativistic electrons in the Earth’s outer radiation belt by parameters of the solar wind and the interplanetary magnetic field measured at the libration point and by the values of the geomagnetic indices. Different adaptive methods are used (namely, artificial neural networks, group method of data handling, and projection to latent structures). The comparison of quality indicators of predictions with a horizon of 1–12 h between each other and with the trivial model prediction has shown that the best result is obtained for the average value of the responses of three neural networks that have been trained with different sets of initial weights. The prediction result of the group method of data handling is close to the result of neural networks, and the projection to latent structures is much worse. It is shown that an increase in the prediction horizon from 1 to 12 h reduces its quality but not dramatically, which makes it possible to use these methods for medium-term prediction.

The results of prediction of geomagnetic indexes characterizing the state of the Earth's magnetosphere obtained with the help of artificial neural networks (ANN) for various prediction horizons are presented. The forecasts are based on multivariate time series including the values of the geomagnetic indices themselves, as well as data about the parameters of solar wind and interplanetary magnetic field, during several latest hours.

It is often necessary to make time series (TS) predictions for several values of the prediction horizon. Usually such predictions are made in autonomous mode, i.e. separately for each horizon value. Meanwhile, it is also possible to make simultaneous predictions for all the desired horizons, or group prediction for several horizons at once. In the preceding studies [1], it has been demonstrated that group determination of parameters in solving multi-parameter inverse problem with a multi-layer perceptron (MLP) may outperform autonomous determination if the approximated dependences of the grouped parameters on the input features of the problem are similar and if the sets of significant input features largely intersect. Last year it has been demonstrated, that the effect also holds for MLP TS prediction with multiple horizons [2]. In the present study, efficiency of group prediction of TS with MLP has been checked at the example of TS of electron daily fluence in near-Earth space, which is characterized by rapid degradation of prediction quality with increasing horizon. Relativistic electrons (RE) of the outer Earth's radiation belt are sometimes called "killer electrons" since they can damage electronic components, resulting in temporary or even complete loss of spacecraft. Daily fluence is summary daily flux of these electrons; at geosynchronous orbit of about 35,000km altitude it is of interest due to the large number of satellites populating this region, and it is predictable thanks to long TS of experimental data available. For this problem, group prediction with average size of groups proved to outperform autonomous and simultaneous prediction. Thus, the positive effect of group determination of outputs in multi-output problem has been confirmed as a property of MLP as data processing algorithm. This study has been performed at the expense of Russian Science Foundation, project no. 16-17-00098.

Prediction of the time series of relativistic electrons flux in the outer radiation belt of the Earth is a complicated task, due to complexity and nonlinearity of the system “solar wind - the Earth’s magnetosphere”. However, using artificial neural networks it is possible to predict the value of the electron flux several hours ahead, based on the hourly time series of electron flux, parameters of solar wind and interplanetary magnetic field. The purpose of this study was to check, which approach provided higher precision of prediction with various horizons from one to twelve hours: autonomous prediction for each of the 12 prediction horizons, or simultaneous prediction for several horizons. An explanation of the obtained results is suggested.