This paper addresses the critical issue of power quality degradation in modern power systems caused by the proliferation of non-linear loads. The study focuses on the development and verification of a parallel-type filter-compensating converter (FCC) specifically designed for three-phase three-wire networks, where the absence of a neutral wire complicates the mitigation of zero-sequence components and current balancing. The authors propose a control strategy based on dividing the network period into twelve characteristic intervals, establishing precise switching sequences for energy consumption and generation modes. A comprehensive simulation model was implemented in the MATLAB/Simulink environment using the Simscape Electrical library to evaluate the converter's dynamic performance. The power circuit parameters, including input chokes and DC-link capacitance, were calculated to ensure stability and minimal voltage ripple. Results from the simulation demonstrate that the FCC effectively transforms the highly non-linear load current (with an initial Total Harmonic Distortion of 35.13%) into a near-sinusoidal network current with a THD of 7.65%. The findings confirm that the proposed converter and its control algorithms successfully ensure electromagnetic compatibility and reactive power compensation, providing a robust solution for industrial three-wire electrical systems.
Abstract–The paper presents the results of an experimental study of the acoustic characteristics of modular MEMS microphones INMP441 and SPH0645, aimed at their application for railway noise monitoring. The experimental methodology and the technical features of the measurement setup in an anechoic chamber are described. The signal processing approach is based on a relative comparison of the signals from the tested microphones with those of a reference microphone and on estimating the power spectral density using the Welch method in MATLAB, which is then used to construct amplitude-frequency responses. The main results of the study include the obtained amplitude-frequency characteristics of the investigated microphones in the 20–20000 Hz frequency range.
This study is devoted to the development of a complex method for extracting the fetal electrocardiogram (fECG) from non-invasive abdominal recordings and for the automated classification of fetal cardiac arrhythmias using machine learning techniques. The early diagnosis of fetal arrhythmias is challenged by a low signal-to-noise ratio in abdominal signals due to the dominance of maternal cardiac activity and the presence of strong physiological interference. To solve this problem, a combined approach integrating multichannel statistical signal separation and spectral analysis of heart rate variability is proposed. The method consists of two stages. At the first stage, fECG extraction is performed using frequency filtering and blind source separation techniques, including principal component analysis (PCA) and independent component analysis (ICA), combined with iterative suppression of the maternal electrocardiogram based on singular value decomposition (SVD). At the second stage, sequences of RR intervals are derived and transformed into a hybrid feature vector combining time-domain heart rate variability metrics and spectral Burg autoregressive coefficients. Numerical results showed that an autoregressive model order of p = 6 provided the best classification performance without indications of overfitting. For the task of binary classification of fetal states (normal or arrhythmia), a unified classification model was developed based on a Support Vector Machine (SVM) classifier combined with a patient-level voting strategy. The proposed method was validated using publicly available datasets: the Abdominal and Direct Fetal ECG Database (ADFECGDB) for assessing signal extraction quality and the Non-Invasive Fetal ECG Arrhythmia Database (NIFEADB) for training and testing the classifier. Evaluated using a Leave-One-Patient-Out (LOPO) cross-validation scheme, the proposed approach achieved an F1-score of 83,3%. Future work will be focused on expanding the diagnostic capabilities of the system by analyzing a broader spectrum of fetal arrhythmias, transitioning to multiclass classification, and applying synthetically generated data to overcome the shortage of annotated fetal pathological records.
Purpose. The study is intended to explore the usability and effectiveness of the acoustic environment safety estimation method in relation to potentially triggering stimuli for people with post-traumatic stress disorder. The proposed method merges power spectral density (PSD) and mel-frequency cepstral coefficient (MFCC) feature extraction with statistical testing and lightweight machine-learning classification and aims to minimize distress responses to salient sound events. The method is designed for portable embedded platforms with optional cloud-side analytics, ensuring analysis efficiency while allowing application by technical and medical personnel within a structured acquisition protocol. Material & Methods. Captured audio recordings were grouped into three categories: non-harmful everyday sounds, harm-related (danger-associated) acoustic stimuli relevant to populated environments and elevated threat salience. The signals were analyzed using Welch-based PSD across 20 Hz–20 kHz (100 Hz bands; Hamming window) and MFCC features (13 cepstral coefficients plus log frame energy). Aggregated feature vectors were computed for each file. The observed data points were divided into three groups of 252 each (N = 756 total). To assess normality, in addition to graphical review, the Shapiro–Wilk test (null hypothesis rejected when p < 0.05) and the Kolmogorov–Smirnov test (null hypothesis rejected when p < 0.20) were used. Parametric (Student’s t) or nonparametric (Mann–Whitney U) pairwise procedures were applied where appropriate. To evaluate differences across the groups, an omnibus Kruskal–Wallis test was used. Classification experiments employed two-layer neural networks, with 10% of the data used for testing. Results. The difference was statistically significant between the three groups of signals for the aggregated PSD- and MFCC-derived descriptors as a result of the proposed method implementation. The use of selected parametric tests, rank-based methods, and conditional nonparametric comparisons was justified by the non-normal distribution of per-file results for both feature families under the study’s testing framework. As illustrated by the study’s compliance matrices for PSD and MFCC inputs, a neural network based on engineered features generated classification consistent with those suggested by spectral and cepstral summaries. Conclusion. The proposed method is internally consistent within the conducted statistical analysis framework for the three-group setup and demonstrates the ability to differentiate between the analyzed acoustic contexts. Moreover, it provides easy implementation within a portable measurement concept and higher analysis speed compared to manual waveforms review. Additionally, integration of audio descriptors with formal hypothesis testing and learned models offers a reliable practical foundation for further testing on independent cohorts and hardware prototype development.
In the article the problem of identification the shading structure of solar panels is considered. The relevance of the study is conditioned by the necessity of rapid and adequate reconfiguration of the connection of photovoltaic modules in order to minimize power losses and overcome the problem of local energy maxima. The object of the study is the spatio-temporal distribution of shading over the surface of a photovoltaic panel. A method for structural decomposition of the illumination field based on the theory of set cardinality in the form of cardinal number vectors is proposed, which makes it possible to formalize the description of shadows and partial shadings (boundary transitions) by coordinates through discretization of the generated energy levels of the cells. To illustrate the method, an example of estimating the geometric (Euclidean) distance between normalized vectors of real shading and a set of five basic fractal functions (Front, Cantor-like, Soft-Cascade, Takagi-like, Stochastic-fBm) is presented, which demonstrates the procedure of shading factorization along the coordinate axes. The obtained results are integrated into a final two-dimensional mathematical function for illumination modeling taking into account weighting coefficients. The proposed approach provides a basis for accelerated decision-making regarding the optimal reconfiguration of solar panels in a quasistatic mode.
