Advances in machine learning (ML) and deep learning (DL) have led to automated sleep staging approaches that achieve high accuracy but often require extensive computational resources and/or highdensity electroencephalograms (EEG).This paper presents a method for sleep staging using features extracted via the Discrete Wavelet Transform (DWT) and Power Spectral Density (PSD), followed by the Gradient Boosting (GB) classifier.The study employs a private dataset and the sleep-EDF dataset, comprising EEG and electrooculograms (EOG) channels.The analysis includes configurations with varying numbers of subjects (75, 20, and 12), and the results demonstrate that the proposed method achieves competitive performance with existing approaches that use complex DL architectures, even with fewer subjects.Feature importance analysis highlights the importance of detail coefficients from DWT and PSD-based features from EEG signals.The findings suggest that simplified methods using low-density EEG and EOG with wellselected features and GB classification can offer a viable alternative to DL approaches for sleep staging.

A high-performance real-time brain-computer interface system capable of identifying dreams has potential for healthcare applications.To address this, we use electroencephalogram (EEG) data from non-rapid eye movement sleep to classify dream experience and noexperience.Using 58 EEG channels, we achieve an accuracy of 0.94, an AUROC of 0.91, and a kappa score of 0.84, accomplished by first filtering the data through multivariate empirical mode decomposition followed by a combination of principal component analysis and common spatial patterns for feature extraction and K-nearest neighbors classifier.Interestingly, comparable results are obtained using 29 or 10 EEG channels selected by permutation-based channel selection.

The development of Brain-Computer Interfaces (BCIs) has been constrained by a predominant focus on signal classification.This paper rather emphasizes the integration of neurophysiological principles, BCI paradigm selection, and rigorous experimental design.By addressing common pitfalls in Machine Learning implementation, we provide researchers with a tutorial and robust framework for BCI development, promoting reproducibility and rigor.Furthermore, by tackling challenges at the intersection of BCI and Machine Learning, this work contributes to the advancement of practical, real-time BCI applications.

Brain-computer Interfaces (BCIs) have been developed towards enhancing communication and control in individuals with motor disabilities and assist in motor rehabilitation, where motor imagery (MI), the mental visualization of limb movement, has been broadly explored.Traditionally, MI-based BCIs utilize electroencephalographic (EEG) recordings to discriminate between limbs motor imagination.This involves applying feature extraction and classification, primarily analyzing signals recorded at the scalp.Despite the success of the traditional sensor space analysis, recent studies have demonstrated that incorporating EEG source imaging (ESI) has led to an improvement of the classification performance.This work studies pipelines on both sensor and source space for classifying upper limb MI.Here, we introduce the use of source average power for the integration of ESI into MI-based BCIs.Our results suggest a significant accuracy improvement of 10% when applying source space analysis with average power against traditional sensor space analysis.This demonstrates that a shift from sensor space analysis to source space analysis can be beneficial for MI classification.

Equipping brain-computer interfaces with dream decoding capabilities could be vital in healthcare applications.We used high-density electroencephalogram data from non-rapid eye movement sleep to conduct qualitative analysis employing multivariate empirical mode decomposition and power spectral density (PSD) for preprocessing and machine learning algorithms to distinguish between a dream experience and no experience.Qualitative analysis shows differences between the two classes, especially in the theta and beta bands.We achieve a classification performance of 0.915 in accuracy, 0.851 in AUROC, and 0.715 in kappa with PSD features and extreme gradient boosting classifier.

Visual inspection of Polysomnography (PSG) recordings by sleep experts, based on established guidelines, has been the gold standard in sleep stage classification. This approach is expensive, time-consuming, and mostly limited to experimental research and clinical cases of major sleep disorders. Various automatic approaches to sleep scoring have been emerging in the past years and are opening the way to a quick computational assessment of sleep architecture that may find its way to the clinics. With the hope to make sleep scoring a fully automated process in the clinics, we report here an ensemble algorithm that aims at not only predicting sleep stages but of doing so with an optimized minimal number of EEG channels. For that, we combine a genetic algorithm-based optimization with a classification framework that minimizes the number of channels used by the machine learning algorithm to quantify sleep stages. This resulted in a sleep scoring with an F1 score of 0.793 for the fully automated model and 0.82 for the model trained on 10 percent of the unseen subject, both with only 3 EEG channels. The ensemble algorithm is based on a combination of extremely randomized trees and MiniRocket classifiers. The algorithm was trained, validated, and tested on night sleep PSG data collected from 7 subjects. Our approach’s novelty lies in using the minimum information needed for automated sleep scoring, based on a systematic search that concurrently selects the optimal-minimum number of EEG channels and the best-performing features for the machine learning classifier. The optimization framework presented in this work may enable new flexible designs for sleep scoring devices suited to studies in the comfort of homes, easily and inexpensively. In this way facilitate experimental and clinical studies in large populations.

The most widely used visual brain-computer interfaces (BCIs) are based on flickering stimulators. It remains unclear whether a passive visual stimulator could be used. Colors are an integral part of everyday life. We analyzed the potential use of both imagined colors and color exposure to control BCIs. We present here a comprehensive study of the feasibility of using the recognition of both exposed and imagined colors from the EEG signals recorded in a previous study. This constitutes the first step towards the development of a color-based BCI. We processed and analyzed the EEG signals from seven subjects recorded whilst they were exposed to three colors (red, green and blue) and whilst they imagined the same colors. The outcomes obtained suggest that the simultaneous use of these two tasks would facilitate the effective recognition of both imagined and displayed colors for the control of a BCI. An analysis of each task separately revealed that color exposure provided more information than color imagination for discriminating between the three colors.

Two different single-phase synchronisation schemes based on orthogonal system generation (OSG) are modelled in the linear, time-periodic framework. The frequency adaptive Hilbert phase locked loop (PLL) and the frequency-fixed second-order-generalised-integrator (SOGI) PLL fit smoothly into existing models for the synchronous reference frame PLL, due to the linear and decoupled nature of the orthogonal system generation. The harmonic transfer functions are established, and shown to accurately capture the frequency coupling properties of the synchronisation techniques. This modelling approach is a milestone towards the ultimate goal of assessing the impact of synchronisation on single-phase converter small-signal stability. The analytic models are compared to simulations, through use of a chirp input signal instead of the typical frequency sweeping technique. The conceptual differences between the Hilbert PLL and SOGI PLL are highlighted, yet without attempts at asserting superiority of either scheme.

This paper investigates the instantaneous properties of voltage and frequency of Voltage Source Converters (VSCs) when they are controlled by Virtual Oscillator Controllers (VOC) in a distorted island grid. Under time-averaged measurements, the VOC exhibits well known linear droop characteristics, which make them a good candidate for control in island grids. However, the instantaneous properties of VOC are influenced by the nonlinear dynamics of the Van der Pol oscillator and provide better synchronization properties under distorted voltage conditions compared to the conventional droop controller. This is shown in a simulation study of parallel operation of two inverters serving a non-linear load that reveals a more stable synchronization at the expense of voltage quality in the case of the VOC compared to the conventional droop controller. These results provide convincing evidence for the adoption of a more complex controller as the VOC in island grids that will naturally be more vulnerable to voltage distortions. At the same time, these results encourage further explorations into other potential benefits of VOC in island grids.

Safety-critical power conversion systems play a major role in the paradigm shift towards more electric aircraft (MEA) architectures. This paper reviews the electrical machines and their power electronic systems that are currently competing in the application of integrated starter-generators (S/Gs) in MEA power systems. Motivated by the strict requirements of sufficient electrical starting capability, super-high power density and ultra-high reliability, additional considerations on the overall system design are necessary, including the power electronic converters (PECs) and integrated thermal designs. These aspects are discussed not only in the light of their many benefits but also of the challenges introduced by the continuous advancements and emerging innovations in the power conversion technology. In achieving the MEA goals and capitalize on all potential benefits, optimization-based design approaches will be necessary, where the aggregation of electric machines, PECs and the aircraft grid is considered as an integrated system to be optimized. This review highlights the importance of these aspects and offers a view on future perspectives and open issues.

This paper proposed a non-filter position sensorless control (PSC) for the permanent-magnet synchronous motor (PMSM), which is based on an a-β frame complex PI controller. The chattering of estimated Back-EMFs is unavoidable with a sliding mode PSC. A low pass filter (LPF) is always employed to smooth Back-EMFs; however phase lag of LPF becomes significant as motor speed increases. A complex PI controller is proposed to estimate back-EMFs without LPFs. The proposed method is implemented by replacing SMC and cascaded LPFs with a single complex PI controller, therefore complexities of proposed PSC have not been increased. Experiments were carried out on a PMSM drive to confirm the advantages of the proposed method over sliding mode PSC.

This paper presents a comparative eigenvalue analysis of the stability characteristics and small signal dynamics of four different control strategies for Synchronous Machine Emulation (SME) by power electronic converters, considering a Synchronous Machine (SM) as the benchmark system. The four SME techniques are selected to represent the most established general approaches for emulating the inertial characteristics of SMs in the control of power electronic converters. The small-signal stability assessment is based on the analysis of system eigenvalues, including evaluation of participation factors and parametric sensitivities. All the investigated techniques can be tuned to obtain similar inertial dynamics under grid frequency variations, but exhibit differences in other small-signal characteristics due to the distinct control system implementations. Among the analyzed cases, the current-controlled virtual synchronous machine has the highest damping of the most oscillatory mode. However, the study shows that the most oscillatory modes of the other techniques are associated with the LCL impedance, and could be further attenuated by active damping techniques.

Current problems related to high-level security access are increasing, leaving organizations and persons unsafe. A recent good candidate to create a robust identity authentication system is based on brain signals recorded with electroencephalograms (EEG). In this paper, EEG-based brain signals of 56 channels, from event-related potentials (ERPs), are used for Subject identification. The ERPs are from positive or negative feedback-related responses of a P300-speller system. The feature extraction part was done with empirical mode decomposition (EMD) extracting 2 intrinsic mode functions (IMFs) per channel, that were selected based on the Minkowski distance. After that, 4 features are computed per IMF; 2 energy features (instantaneous and teager energy) and 2 fractal features (Higuchi and Petrosian fractal dimension). Support vector machine (SVM) was used for the classification stage with an accuracy index computed using 10-folds cross-validation for evaluating the classifier's performance. Since high-density EEG information was available, the well-known backward-elimination and forward-addition greedy algorithms were used to reduce or increase the number of channels, step by step. Using the proposed method for subject identification from a positive or negative feedback-related response and then identify the subject will add a layer to improve the security system. The results obtained show that subject identification is feasible even using a low number of channels: E.g., 0.89 of accuracy using 5 channels with a mixed population and 0.93 with a male-only population.

Conventional small-signal approaches present obvious advantages applied in analyzing closed-loop stability in power electronics systems. However, linearized models fall short of predicting any nonlinear behavior which means not able to acquire the accurate stability margin. This paper mainly analyzes bifurcation behaviors of the Modular Multilevel Converter (MMC) connected to a weak grid and calculates the parameters' precise stability margin. The nonlinear continuous-time averaging model of the MMC is established to study two common bifurcation phenomena, i.e. Saddle-Node bifurcation (SNB) and Hopf bifurcation (HB). Characteristics of limit cycles caused by HB are analyzed by changing the value of a certain parameter. The theoretical analysis has been validated by the time domain simulation results.

The impedance modeling of modular multilevel converters (MMCs) is a prerequisite for applying the impedance-based stability analysis method to an MMC-based high-voltage direct current (MMC-HVDC) transmission system. The objective of this paper is to develop impedance models of a three-phase MMC in order to assess its stability. The analytical ac-side and dc-side impedance models of the three-phase MMC with and without circulating current control (CCC) are derived separately with consideration of the MMC internal dynamics. In doing that, a control strategy for the MMC in an application of wind power integration through MMC-HVDC is taken into consideration when calculating the ac-side impedances. The analytical impedance models derived for the MMC are verified by the method based on injecting a small-signal perturbation in the detailed nonlinear time-domain simulation model of the MMC.

Harmonic voltages and currents in electrical systems, when present to a certain degree, represent not only a\npower quality problem but they are also strongly associated with the electrical system overall losses and they\nare arguably a source of instability and a safety concern. Mitigating harmonics distortion across the entire\nsystem by actively reducing harmonic currents propagation is an effective way of coping with these issues and\ncan be dealt with the injection of a compensating current waveform with an active filter installed at a given\nbus. This paper shows how, by shaping the compensating current waveform in an optimal way, the overall\nelectrical system harmonic distortion can be optimally reduced in a cost effective manner with a minimum\nsize of the compensating device. The process of shaping this optimal compensating current is shown by how\nits components are defined by the optimization algorithm using the phase and amplitude of each harmonic as\ndegrees of freedom in the process of finding the optimal waveform. A marine vessel distribution grid is used\nas representative example to prove the concept.

Examines the development of marine vehicle power systems from an historical perspective. from the earliest forms of power systems to the current development of all electric ship technologies.

A generalized and versatile approach for control of the circulating currents of Modular Multilevel Converters (MMC) has been recently proposed by using Lagrange Multipliers in the ABC frame. This approach is capable of analytically obtaining the differential current references associated with the desired operating conditions imposed to the optimization procedure. Previous implementations of this control approach have been sensitive to harmonic distortion of the control signals when applied to systems with a low number of levels, and have depended on average value calculations of sinusoidal variables, slowing down the dynamics of the control and making the tuning of the control system difficult under certain conditions. This work demonstrates that the performance of the versatile Lagrange-based control can be improved by introducing adaptive filtering based on Second Order Generalized Integrators (SOGIs) for processing the system variables needed for calculating the circulating current references. In addition to improved control dynamics and easier tuning of system controllers, the use of the adaptive filters provides inherent capability for operation under frequency variations, and is therefore suitable for control of MMCs operating in weak grids as well as for high- or medium-voltage motor drives.

In this paper a matrix converter modulation adapted for the wind energy conversion system of series-connected turbines is presented. The considered conversion system is composed of a permanent magnet synchronous generator, a matrix converter, a high frequency transformer and a diode bridge rectifier for the AC-DC conversion. This conversion system topology allows for a modification, introduced in this paper, of the indirect space vector modulation that increases the current ratio of the matrix converter and thus enhances its advantage for this wind park configuration. The modulation can be further modified into a square wave modulation for high frequency operation of the transformer. An efficiency analysis of the suggested modulation and its high frequency version is performed. Also some attention is given to the flux balance issue in the transformer and a possible solution for a modulation yielding balanced flux is given.

The efficiency and thermal performance of AC-AC converters in series-connected DC wind turbines is the object of this paper. The considered conversion system is composed of a permanent magnet synchronous generator, an AC-AC converter, a high frequency transformer and a diode bridge rectifier. Two different converter topologies are compared: the Back-to-Back converter and matrix converter, in terms of efficiency and thermal performance. The modulations used for each converter are presented. The control for each converter is described as well as the method used for loss and temperature calculations. The results show that the matrix converter is more efficient than the back-to-back converter, however in terms of thermal performance and temperature fluctuation, the back-to-back is slightly superior.

In this paper a centralized stabilizer is designed for a multi-converter dc microgrid. DC distribution systems are gaining more popularity due to high efficiency, easy integration of various sources, and high reliability. These dc systems typically consist of several power electronic converters to link different sources and loads to the grid. The stability of such power electronic-based systems is a major design and control consideration due to potential negative incremental resistance introduced by power electronic converters. In other terms, the power converters act as Constant Power Loads (CPLs), which can drive the system to instability. This paper, after detail stability analysis of a multi-converter dc system, sets forth a centralized control of the power converters that improves the stability margins. Simulation results confirm the stability analysis and performance of the control method.

This work presents a simple approach to eliminate DC power ripple in a three-phase Modular Multilevel Converter (MMC) while operating with balanced (load) currents during unbalanced grid faults. In this study, the MMC is connected to an AC grid on one of its terminals, and to a constant ideal DC source on the other, while a single phase fault takes place. The presented “first” approach is based on the combined application of a Notch filter and a Resonant PI controller. The main idea is to eliminate the zero-sequence oscillations that occur in the MMC differential currents when the unbalanced fault takes place. This strategy may be applied jointly with a constant circulating current controller on the one hand or with a constant energy per phase (implying an injection of a second harmonic to the differential current), on the other. Moreover, the load currents are kept balanced even during the fault in order to demonstrate the potential application of the MMC as a power oscillation firewall of some sort; i.e., to simultaneously keep constant DC power and balanced load currents at all time.