Electronics

16 pages, 3904 KB

Open AccessFeature PaperEditor’s ChoiceArticle

Common Mode Voltage Elimination for Quasi-Switch Boost T-Type Inverter Based on SVM Technique

by Duc-Tri Do, Minh-Khai Nguyen, Van-Thuyen Ngo, Thanh-Hai Quach and Vinh-Thanh Tran

Electronics 2020, 9(1), 76; https://doi.org/10.3390/electronics9010076 - 1 Jan 2020

Cited by 17 | Viewed by 3934

In this paper, the effect of common-mode voltage generated in the three-level quasi-switched boost T-type inverter is minimized by applying the proposed space-vector modulation technique, which uses only medium vectors and zero vector to synthesize the reference vector. The switching sequence is selected [...] Read more.

In this paper, the effect of common-mode voltage generated in the three-level quasi-switched boost T-type inverter is minimized by applying the proposed space-vector modulation technique, which uses only medium vectors and zero vector to synthesize the reference vector. The switching sequence is selected smoothly for inserting the shoot-through state for the inverter branch. The shoot-through vector is added within the zero vector in order to not affect the active vectors as well as the output voltage. In addition, the shoot-through control signal of active switches of the impedance network is generated to ensure that its phase is shifted 90 degrees compared to shoot through the signal of the inverter leg, which provides an improvement in reducing the inductor current ripple and enhancing the voltage gain. The effectiveness of the proposed method is verified through simulation and experimental results. In addition, the superiority of the proposed scheme is demonstrated by comparing it to the conventional pulse-width modulation technique. Full article

(This article belongs to the Special Issue Power Converters in Power Electronics)

► Show Figures

Figure 1

14 pages, 10153 KB

Open AccessFeature PaperEditor’s ChoiceArticle

Low-Speed Performance Improvement of Direct Torque Control for Induction Motor Drives Fed by Three-Level NPC Inverter

by Samer Saleh Hakami, Ibrahim Mohd Alsofyani and Kyo-Beum Lee

Electronics 2020, 9(1), 77; https://doi.org/10.3390/electronics9010077 - 1 Jan 2020

Cited by 12 | Viewed by 4073

Abstract

Classical direct torque control (DTC) is considered one of the simplest and fastest control algorithms in motor drives. However, it produces high torque and flux ripples due to the implementation of the three-level hysteresis torque regulator (HTR). Although, increasing the level of HTR [...] Read more.

Classical direct torque control (DTC) is considered one of the simplest and fastest control algorithms in motor drives. However, it produces high torque and flux ripples due to the implementation of the three-level hysteresis torque regulator (HTR). Although, increasing the level of HTR and utilizing multilevel inverters has a great contribution in torque and flux ripples reduction, stator flux magnitude of induction motor (IM) droops at every switching sector transition, particularly at low-speed operation. This problem occurs due to the utilization of a zero voltage vector, where the domination of stator resistance is very high. A simple flux regulation strategy (SFRS) is applied for low-speed operation for DTC of IM. The proposed DTC-SFRS modifies the output status of the five-level HTR depending on the flux error, torque error, and stator flux position. This method regulates the stator flux for both forward and reverse rotational directions of IM with retaining the basic structure of classical DTC. By using the proposed algorithm, the stator flux is regulated, hence pure sinusoidal current waveform is achieved, which results in lower total harmonics distortion (THD). The effectiveness of the proposed DTC-SFRS is verified through simulation and experimental results. Full article

(This article belongs to the Special Issue High Power Electric Traction Systems)

► Show Figures

Figure 1

16 pages, 4395 KB

Open AccessFeature PaperEditor’s ChoiceArticle

Development of a Compact, IoT-Enabled Electronic Nose for Breath Analysis

by Akira Tiele, Alfian Wicaksono, Sai Kiran Ayyala and James A. Covington

Electronics 2020, 9(1), 84; https://doi.org/10.3390/electronics9010084 - 1 Jan 2020

Cited by 51 | Viewed by 9410

Abstract

In this paper, we report on an in-house developed electronic nose (E-nose) for use with breath analysis. The unit consists of an array of 10 micro-electro-mechanical systems (MEMS) metal oxide (MOX) gas sensors produced by seven manufacturers. Breath sampling of end-tidal breath is [...] Read more.

In this paper, we report on an in-house developed electronic nose (E-nose) for use with breath analysis. The unit consists of an array of 10 micro-electro-mechanical systems (MEMS) metal oxide (MOX) gas sensors produced by seven manufacturers. Breath sampling of end-tidal breath is achieved using a heated sample tube, capable of monitoring sampling-related parameters, such as carbon dioxide (CO₂), humidity, and temperature. A simple mobile app was developed to receive real-time data from the device, using Wi-Fi communication. The system has been tested using chemical standards and exhaled breath samples from healthy volunteers, before and after taking a peppermint capsule. Results from chemical testing indicate that we can separate chemical standards (acetone, isopropanol and 1-propanol) and different concentrations of isobutylene. The analysis of exhaled breath samples demonstrate that we can distinguish between pre- and post-consumption of peppermint capsules; area under the curve (AUC): 0.81, sensitivity: 0.83 (0.59–0.96), specificity: 0.72 (0.47–0.90), p-value: <0.001. The functionality of the developed device has been demonstrated with the testing of chemical standards and a simplified breath study using peppermint capsules. It is our intention to deploy this system in a UK hospital in an upcoming breath research study. Full article

(This article belongs to the Special Issue Design and Application of Biomedical Circuits and Systems)

► Show Figures

Figure 1

18 pages, 3546 KB

Open AccessFeature PaperEditor’s ChoiceArticle

Face–Iris Multimodal Biometric Identification System

by Basma Ammour, Larbi Boubchir, Toufik Bouden and Messaoud Ramdani

Electronics 2020, 9(1), 85; https://doi.org/10.3390/electronics9010085 - 1 Jan 2020

Cited by 86 | Viewed by 10482

Abstract

Multimodal biometrics technology has recently gained interest due to its capacity to overcome certain inherent limitations of the single biometric modalities and to improve the overall recognition rate. A common biometric recognition system consists of sensing, feature extraction, and matching modules. The robustness [...] Read more.

Multimodal biometrics technology has recently gained interest due to its capacity to overcome certain inherent limitations of the single biometric modalities and to improve the overall recognition rate. A common biometric recognition system consists of sensing, feature extraction, and matching modules. The robustness of the system depends much more on the reliability to extract relevant information from the single biometric traits. This paper proposes a new feature extraction technique for a multimodal biometric system using face–iris traits. The iris feature extraction is carried out using an efficient multi-resolution 2D Log-Gabor filter to capture textural information in different scales and orientations. On the other hand, the facial features are computed using the powerful method of singular spectrum analysis (SSA) in conjunction with the wavelet transform. SSA aims at expanding signals or images into interpretable and physically meaningful components. In this study, SSA is applied and combined with the normal inverse Gaussian (NIG) statistical features derived from wavelet transform. The fusion process of relevant features from the two modalities are combined at a hybrid fusion level. The evaluation process is performed on a chimeric database and consists of Olivetti research laboratory (ORL) and face recognition technology (FERET) for face and Chinese academy of science institute of automation (CASIA) v3.0 iris image database (CASIA V3) interval for iris. Experimental results show the robustness. Full article

(This article belongs to the Special Issue Recent Advances in Biometrics and its Applications)

► Show Figures

Figure 1

28 pages, 9694 KB

Open AccessFeature PaperEditor’s ChoiceArticle

Design of an Intrinsically Safe Series-Series Compensation WPT System for Automotive LiDAR

by Luiz A. Lisboa Cardoso, Vítor Monteiro, José Gabriel Pinto, Miguel Nogueira, Adérito Abreu, José A. Afonso and João L. Afonso

Electronics 2020, 9(1), 86; https://doi.org/10.3390/electronics9010086 - 1 Jan 2020

Cited by 2 | Viewed by 3917

Abstract

The earliest and simplest impedance compensation technique used in inductive wireless power transfer (WPT) design is the series-series (SS) compensation circuit, which uses capacitors in series with both primary and secondary coils of an air-gapped transformer. Despite of its simplicity at the resonant [...] Read more.

The earliest and simplest impedance compensation technique used in inductive wireless power transfer (WPT) design is the series-series (SS) compensation circuit, which uses capacitors in series with both primary and secondary coils of an air-gapped transformer. Despite of its simplicity at the resonant condition, this configuration exhibits a major sensitivity to variations of the load attached to the secondary, especially when higher coupling coefficients are used in the design. In the extreme situation that the secondary coil is left at open circuit, the current at the primary coil may increase above the safety limits for either the power converter driving the primary coil or the components in the primary circuit, including the coil itself. An approach often used to minimize this problem is detuning, but this also reduces the electrical efficiency of the power transfer. In low power, fixed-distance stationary WPT, a fair trade-off between efficiency and safety must be verified. This paper aims to consolidate a simple design procedure for such a SS-compensation, exemplifying its use in the prototype of a WPT system for automotive light detection and ranging (LiDAR) equipment. The guidelines herein provided should equally apply to other low power applications. Full article

(This article belongs to the Section Power Electronics)

► Show Figures

Figure 1

9 pages, 3402 KB

Open AccessEditor’s ChoiceArticle

A Compact 3.3–3.5 GHz Filter Based on Modified Composite Right-/Left-Handed Resonator Units

by Shanwen Hu, Yunqing Hu, Haiyu Zheng, Weiguang Zhu, Yiting Gao and Xiaodong Zhang

Electronics 2020, 9(1), 1; https://doi.org/10.3390/electronics9010001 - 18 Dec 2019

Cited by 12 | Viewed by 4632

Abstract

In the RF (Radio Frequency) front-end of a communication system, bandpass filters (BPFs) are used to send passband signals and reject stopband signals. Substrate-integrated waveguides (SIW) are widely used in RF filter designs due to their low loss and low cost and the [...] Read more.

In the RF (Radio Frequency) front-end of a communication system, bandpass filters (BPFs) are used to send passband signals and reject stopband signals. Substrate-integrated waveguides (SIW) are widely used in RF filter designs due to their low loss and low cost and the flexibility of their integration properties. However, SIW filters under 6 GHz are still too large to meet the requirement of portable communication devices due to their long wavelength. In this paper, a very compact fully integrated SIW filter is proposed and designed with RT6010 dielectric material to meet the small size requirement of portable devices for next-generation sub-6 G applications. The proposed filter contains two sawtooth-shaped composite right-/left-handed (CRLH) resonator units, instead of traditional rectangular-shaped CRLH resonator units, which makes the filter more compact and cost effective. The filter is designed and fabricated on an RT6010 substrate, with a size of only 10 mm × 7.4 mm. The measurement results illustrated that the proposed BPF shows a passband covering the frequency range of 3.25–3.45 GHz; the minimum passband insertion loss is only 2.4 dB; the stopband rejection is better than −20 dB throughout the frequencies below 3.0 GHz and above 3.8 GHz; S11 is as low as −37 dB at 3.35 GHz; and the group delay variation is only 1.4 ns throughout the operation bandwidth. Full article

(This article belongs to the Special Issue RF/Mm-Wave Circuits Design and Applications)

► Show Figures

Figure 1

18 pages, 5456 KB

Open AccessEditor’s ChoiceArticle

A Deep Learning-Based Scatter Correction of Simulated X-ray Images

by Heesin Lee and Joonwhoan Lee

Electronics 2019, 8(9), 944; https://doi.org/10.3390/electronics8090944 - 27 Aug 2019

Cited by 34 | Viewed by 8018

Abstract

X-ray scattering significantly limits image quality. Conventional strategies for scatter reduction based on physical equipment or measurements inevitably increase the dose to improve the image quality. In addition, scatter reduction based on a computational algorithm could take a large amount of time. We [...] Read more.

X-ray scattering significantly limits image quality. Conventional strategies for scatter reduction based on physical equipment or measurements inevitably increase the dose to improve the image quality. In addition, scatter reduction based on a computational algorithm could take a large amount of time. We propose a deep learning-based scatter correction method, which adopts a convolutional neural network (CNN) for restoration of degraded images. Because it is hard to obtain real data from an X-ray imaging system for training the network, Monte Carlo (MC) simulation was performed to generate the training data. For simulating X-ray images of a human chest, a cone beam CT (CBCT) was designed and modeled as an example. Then, pairs of simulated images, which correspond to scattered and scatter-free images, respectively, were obtained from the model with different doses. The scatter components, calculated by taking the differences of the pairs, were used as targets to train the weight parameters of the CNN. Compared with the MC-based iterative method, the proposed one shows better results in projected images, with as much as 58.5% reduction in root-mean-square error (RMSE), and 18.1% and 3.4% increases in peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM), on average, respectively. Full article

(This article belongs to the Special Issue Deep Neural Networks and Their Applications)

► Show Figures

Figure 1

20 pages, 36563 KB

Open AccessEditor’s ChoiceArticle

Fallen People Detection Capabilities Using Assistive Robot

by Saturnino Maldonado-Bascón, Cristian Iglesias-Iglesias, Pilar Martín-Martín and Sergio Lafuente-Arroyo

Electronics 2019, 8(9), 915; https://doi.org/10.3390/electronics8090915 - 21 Aug 2019

Cited by 45 | Viewed by 8325

Abstract

One of the main problems in the elderly population and for people with functional disabilities is falling when they are not supervised. Therefore, there is a need for monitoring systems with fall detection functionality. Mobile robots are a good solution for keeping the [...] Read more.

One of the main problems in the elderly population and for people with functional disabilities is falling when they are not supervised. Therefore, there is a need for monitoring systems with fall detection functionality. Mobile robots are a good solution for keeping the person in sight when compared to static-view sensors. Mobile-patrol robots can be used for a group of people and systems are less intrusive than ones based on mobile robots. In this paper, we propose a novel vision-based solution for fall detection based on a mobile-patrol robot that can correct its position in case of doubt. The overall approach can be formulated as an end-to-end solution based on two stages: person detection and fall classification. Deep learning-based computer vision is used for person detection and fall classification is done by using a learning-based Support Vector Machine (SVM) classifier. This approach mainly fulfills the following design requirements—simple to apply, adaptable, high performance, independent of person size, clothes, or the environment, low cost and real-time computing. Important to highlight is the ability to distinguish between a simple resting position and a real fall scene. One of the main contributions of this paper is the input feature vector to the SVM-based classifier. We evaluated the robustness of the approach using a realistic public dataset proposed in this paper called the Fallen Person Dataset (FPDS), with 2062 images and 1072 falls. The results obtained from different experiments indicate that the system has a high success rate in fall classification (precision of 100% and recall of 99.74%). Training the algorithm using our Fallen Person Dataset (FPDS) and testing it with other datasets showed that the algorithm is independent of the camera setup. Full article

(This article belongs to the Special Issue Machine Learning Techniques for Assistive Robotics)

► Show Figures

Graphical abstract

18 pages, 3513 KB

Open AccessEditor’s ChoiceArticle

Multivariate Temporal Convolutional Network: A Deep Neural Networks Approach for Multivariate Time Series Forecasting

by Renzhuo Wan, Shuping Mei, Jun Wang, Min Liu and Fan Yang

Electronics 2019, 8(8), 876; https://doi.org/10.3390/electronics8080876 - 7 Aug 2019

Cited by 297 | Viewed by 28425

Abstract

Multivariable time series prediction has been widely studied in power energy, aerology, meteorology, finance, transportation, etc. Traditional modeling methods have complex patterns and are inefficient to capture long-term multivariate dependencies of data for desired forecasting accuracy. To address such concerns, various deep learning [...] Read more.

Multivariable time series prediction has been widely studied in power energy, aerology, meteorology, finance, transportation, etc. Traditional modeling methods have complex patterns and are inefficient to capture long-term multivariate dependencies of data for desired forecasting accuracy. To address such concerns, various deep learning models based on Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN) methods are proposed. To improve the prediction accuracy and minimize the multivariate time series data dependence for aperiodic data, in this article, Beijing PM2.5 and ISO-NE Dataset are analyzed by a novel Multivariate Temporal Convolution Network (M-TCN) model. In this model, multi-variable time series prediction is constructed as a sequence-to-sequence scenario for non-periodic datasets. The multichannel residual blocks in parallel with asymmetric structure based on deep convolution neural network is proposed. The results are compared with rich competitive algorithms of long short term memory (LSTM), convolutional LSTM (ConvLSTM), Temporal Convolution Network (TCN) and Multivariate Attention LSTM-FCN (MALSTM-FCN), which indicate significant improvement of prediction accuracy, robust and generalization of our model. Full article

(This article belongs to the Special Issue Deep Learning Applications with Practical Measured Results in Electronics Industries)

► Show Figures

Figure 1

34 pages, 364 KB

Open AccessEditor’s ChoiceReview

Machine Learning Interpretability: A Survey on Methods and Metrics

by Diogo V. Carvalho, Eduardo M. Pereira and Jaime S. Cardoso

Electronics 2019, 8(8), 832; https://doi.org/10.3390/electronics8080832 - 26 Jul 2019

Cited by 1292 | Viewed by 70677

Abstract

Machine learning systems are becoming increasingly ubiquitous. These systems’s adoption has been expanding, accelerating the shift towards a more algorithmic society, meaning that algorithmically informed decisions have greater potential for significant social impact. However, most of these accurate decision support systems remain complex [...] Read more.

Machine learning systems are becoming increasingly ubiquitous. These systems’s adoption has been expanding, accelerating the shift towards a more algorithmic society, meaning that algorithmically informed decisions have greater potential for significant social impact. However, most of these accurate decision support systems remain complex black boxes, meaning their internal logic and inner workings are hidden to the user and even experts cannot fully understand the rationale behind their predictions. Moreover, new regulations and highly regulated domains have made the audit and verifiability of decisions mandatory, increasing the demand for the ability to question, understand, and trust machine learning systems, for which interpretability is indispensable. The research community has recognized this interpretability problem and focused on developing both interpretable models and explanation methods over the past few years. However, the emergence of these methods shows there is no consensus on how to assess the explanation quality. Which are the most suitable metrics to assess the quality of an explanation? The aim of this article is to provide a review of the current state of the research field on machine learning interpretability while focusing on the societal impact and on the developed methods and metrics. Furthermore, a complete literature review is presented in order to identify future directions of work on this field. Full article

(This article belongs to the Section Artificial Intelligence)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Editor’s Choice Articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI