This paper refers to the method of using the deep neural long-short-term memory (LSTM) network for the problem of electrocardiogram (ECG) signal classification. ECG signals contain a lot of subtle information analyzed by doctors to determine the type of heart dysfunction. Due to the large number of signal features that are difficult to identify, raw ECG data is usually not suitable for use in machine learning. The article presents how to transform individual ECG time series into spectral images for which two characteristics are determined, which are instantaneous frequency and spectral entropy. Feature extraction consists of converting the ECG signal into a series of spectral images using short-term Fourier transformation. Then the images were converted using Fourier transform again to two signals, which includes instantaneous frequency and spectral entropy. The data set transformed in this way was used to train the LSTM network. During the experiments, the LSTM networks were trained for both raw and spectrally transformed data. Then, the LSTM networks trained in this way were compared with each other. The obtained results prove that the transformation of input signals into images can be an effective method of improving the quality of classifiers based on deep learning.

Context. We investigate the wave heating problem of a solar quiet region and present its plausible solution without involving shock formation. Aims. We aim to use numerical simulations to study wave propagation and dissipation in the partially ionized solar atmosphere, whose model includes both neutrals and ions. Methods. We used a 2.5D two-fluid model of the solar atmosphere to study the wave generation and propagation. The source of these waves is the solar convection located beneath the photosphere. Results. The energy carried by the waves is dissipated through ion-neutral collisions, which replace shocks used in some previous studies as the main source of local heating in quiet regions. Conclusions. We show that the resulting wave dissipation is sufficient to balance radiative and thermal energy losses, and to sustain a quasi-stationary atmosphere whose averaged temperature profile agrees well with the observationally based semi-empirical model of Avrett & Loeser (2008, ApJS, 175, 229).

We perform numerical simulations of acoustic waves in a two-fluid model of quiet region of the solar atmosphere. The two-fluid model describes partially ionized (non-magnetized) solar plasma, whose main components are neutral atoms, protons, and electrons. The waves are excited by a monochromatic driver, which operates at the bottom of the solar photosphere. Our numerical results show that the driver excites ion and neutral acoustic waves whose propagation is affected by the gravity. As a result, the acoustic waves with periods higher than a local acoustic cut-off period are evanescent, while lower waveperiods are free to reach the solar corona. Acoustic waves, which are evanescent in the photosphere and low chromosphere, tunnel their energy into the upper chromosphere and the transition region. The wave propagation to the solar corona is affected by partial wave reflection that occurs in the transition region, and is responsible for formation of a cavity, where the waves are trapped. Fourier power analysis of temporal characteristic of plasma quantities reveals that a spectrum of various periods is generated. While oscillations traced in ion and neutral velocities look very similar, dynamics of mass densities of ions and neutrals differs a lot. The obtained results clearly show that the two-fluid model provides new insights into the acoustic wave propagation in a more realistic (partially ionized) quiet region of the solar atmosphere.

This paper describes the method developed using the Extreme Gradient Boosting (Xgboost) algorithm that allows high-resolution imaging using the ultrasound tomography (UST) signal. More precisely, we can locate, isolate, and use the reflective peaks from the UST signal to achieve high-resolution images with low noise, which are far more useful for the location of points where the reflection occurred inside the experimental tank. Each reconstruction is divided into two parts, estimation of starting points of wave packets of raw signal (SAT—starting arrival time) and image reconstruction via XGBoost algorithm based on SAT matrix. This technology is the basis of a project to design non-invasive monitoring and diagnostics of technological processes. In this paper, we present a method of the complete solution for monitoring industrial processes. The measurements used in the study were obtained with the author’s solution of ultrasound tomography.

With the use of our JOANNA code, which solves radiative equations for ion + electron and neutral fluids, we perform realistic 2.5D numerical simulations of plasma outflows associated with the solar granulation. These outflows exhibit physical quantities that are consistent, to the order of magnitude, with the observational findings for mass and energy losses in the upper chromosphere, transition region, and inner corona, and they may originate the fast solar wind.

Despite numerous observational and theoretical attempts, the heating problem of the solar chromosphere still remains unsolved. We develop a novel 3D two-fluid model that accounts for dynamics of charged species and neutrals, and use it to perform the numerical simulations of granulation driven jets and associated waves in a quiet region of the solar chromosphere. The energy carried by the waves is dissipated through ion–neutral collisions, which are sufficient to balance radiative energy losses and to sustain the quasi-stationary atmosphere whose ion and neutral number densities, ionization fraction, and temperature profiles are relatively close to the observationally based semi-empirical model. Additional verification of our results is provided by a good fit of the numerically predicted waveperiod variations with height to the recent observational data. These observational validations of the numerical results demonstrate that the wave heating problem of a quiet region of the chromosphere may be solved.