Search

Accurately estimating the capacity of lithium-ion batteries in electric vehicles (EVs) is critical for making correct management decisions. However, the randomness of the charging voltage range of EVs can lead to missing observations or reduced accuracy of capacity estimation methods. This paper proposes an adaptive battery capacity estimation method suitable for arbitrary charging voltage range based on incremental capacity (IC) analysis and data-driven techniques. All charging conditions of EVs are divided into three categories according to the charging voltage range. Three data-driven estimation sub-methods with sequential application priority are designed for the three charging conditions separately, including back-propagation neural network with IC peak coordinates as input, ensemble learning with local high IC curve as input, and linear regression with ampere-hour coordinate transformation. The method is based on a priori knowledge to select a suitable estimation sub-method under different charging conditions, so as to improve the adaptability. Experimental data is collected from eight commercial lithium-ion battery modules for model establishment and verification. Over 250,000 experimental samples at different states of health and random charging ranges show that the method can accurately estimate battery capacity under arbitrary charging conditions, with a maximum error of 2%.

The microgrid (MG) which effectively utilizes distributed energy resources (DERs) is crucial to the modern power system. However, since the islanded MG with high penetration of DERs inherently lacks sufficient inertial support, the frequency regulation is challenging for the voltage source converter (VSC)-based MG. This paper presents a cascaded model predictive control (MPC) scheme for both the outer and inner loops of primary control, which improves frequency regulation capability with the advantage of satisfactory dynamic response and high tracking accuracy. In the outer loop, to enhance dynamic frequency characteristics, the MPC-based virtual synchronous generator (VSG) method with two control objectives is proposed. According to the movement direction of frequency under different load switching cases, the corresponding frequency response process speeded up or slowed down properly. In the inner loop, an improved finite-set double vector MPC (DV-MPV) is proposed to follow the output reference from the outer loop, which contributes to accurate frequency control by reducing tracking error, and faster dynamic response as well. The simulation and experimental results further demonstrated the effectiveness of the proposed method.

Three-level inverter has been widely used in electric drives and renewable power generation system. However, the common mode voltage (CMV) and output current ripple reduction of three-level inverter should be carefully regulated for the appropriate operation. As these two targets are mutually interrelated, the conventional space vector modulation (SVM) scheme cannot cope with them decently. To overcome this drawback, this paper proposes a novel modified SVM scheme to realize CMV and current ripple reduction simultaneously. The generating mechanism of output current ripple is analyzed, and the relationship between current ripple and utilized vectors is quantitatively described. Based on the analyses, the proposed optimal current ripple and CMV reduction pulse width modulation (OCR-CMVRPWM) method adopts four voltage vectors (large, medium, small and zero vectors) to reduce root mean square (RMS) value of current ripple by allocating the sequence and dwell times of utilized vectors. Compared to the existing PWM methods, the proposed OCR-CMVRPWM scheme can reduce the peak of CMV to half of value using the conventional SVM, and obtain a significant reduction of output current total harmonic distortion (THD). The effectiveness of the proposed method is validated by both simulation and experimental results.

In a multi-tenant business environment, tenants share the same applications and databases to store their data. Due to the widespread use of a multi-tenant environment, the service providers face difficult challenges daily. These challenges are condensed in how to guarantee the quality of service provided to tenants, which are documented in a formal document known as a Service Level Agreement (SLA). In addition, SLA should consider the irregular patterns of workload of tenants which may affect the level of guarantee. In this research, an Enhanced Multi-Tenant Database Management System (EMT DBMS) is proposed. In addition, an Enhanced Multi-tenant Migration Algorithm called EMT-M is presented, which aims to migrate the violated tenants depending on both the number of SLA violations and the variance rate. Experimental results prove that the proposed EMT-M algorithm is ideal for migrating violated tenants, as it reduces the number of SLA violations compared to previous migration algorithms.

With the rapid adoption of multi-tenant databases, the cloud provider consolidates multiple tenants’ database on server machines, where the tenants share a common application and database instances. To ensure the quality of service (QoS) for the leased resources, both sides (i.e., the user and the provider) create a Service Level Agreement (SLA). Higher SLA violations result in high SLA contractual penalties and increase the possibility of losing the tenant. In addition, the unusual workload patterns of each tenant transactions require seamless adjustments due to the sudden burden changes and variability. As a result, to satisfy simultaneously availability and performance tenant requirements, it is necessary to perform reliable tenant migration and replication to distribute the workload to a flexible set of sites and avoid SLA violations. In this research, a cluster-based multi-tenant database management system (CB-MT DBMS) is proposed, which takes the migration and replication decisions in advance by monitoring and acting before the violation of the SLA occurs. In addition, a dynamic proactive multi-tenant database migration and replication MTDB-MR algorithm is proposed to reduce collisions and inconsistencies between migration and replication decisions for a group of violated tenants. Experimental results show that the proposed MTDB-MR algorithm is the ideal candidate for migration and replication of the violated multi-tenant databases, as it minimizes the total number of SLA violations, the number of multi-tenant clients SLA violations, client sites average response time and total execution time of each multi-tenant client site as compared to the previous algorithms.

The representation system in modern radars is carried out by obtaining digital signals. In the case of tracking radars, the representation is updated in the order of tenths of a millisecond. In order to achieve to update the display in real time, the acquisition, processing and representation process must comply with the time requirement imposed by the radar. The objective of this work is the visualization of the information for a medium of this type in real time using the Odroid XU4 board. For their solution, the parallel programming method was used, through the creation of threads in the Qt Creator Integrated Development Environment and the use of the "cauce" segmentation parallel programming pattern. The result allowed an efficient use of computing resources, obtaining a decrease of the execution time and a greater acceleration with respect to the sequential variant., El sistema de representación en los radares modernos se realiza a partir de la obtención de señales digitales. En el caso de los radares de seguimiento la actualización de la representación se realiza en el orden de las décimas de milisegundos. Para lograr la actualización del display en tiempo real el proceso de adquisición, procesamiento de la señal y representación tiene que cumplir con la exigencia de tiempo de exploración del radar. El objetivo del trabajo es lograr la visualización de la información para un radar en tiempo real empleando la placa Odroid XU4. Se utilizó el método de programación paralela, mediante la creación de hilos en el Entorno de Desarrollo Integrado Qt Creator y el empleo del patrón de programación paralela segmentación de cauce. Esto permitió el uso eficiente de los recursos de cómputo, obteniéndose una disminución del tiempo de ejecución y una mayor aceleración con respecto a la variante secuencial.

Arabic datasets that have two or more records for the same world entity (i.e. person, object, etc.) make institutions suffer from low quality and degraded performance due to duplication in their Arabic datasets without having any mechanism for detecting these duplicates. The operation that distinguishes records for the same real-world entity is called Entity Resolution (ER). It is considered as a tool for linking records across databases as well as for matching query records with existing databases in real-time. Indexing is a major step in the ER process that aims at reducing the search space. Several indexing techniques are available for use with the ER process in general for English Databases. However, such techniques are not validated if they work well with other languages, such as Arabic. The Dynamic Similarity Aware Inverted Index (DySimII) is one of the indexing techniques that are utilized with dynamic databases to match query records in real time and is demonstrated to work well with English language. In this paper, we propose a framework—Arabic Real Time Entity Resolution (ARTER)—that uses DySimII with Arabic databases to perform real time ER. We also examine using different string similarity functions required for comparing records in the matching process for the aim of evaluating which similarity function is more suitable for comparing Arabic strings. A real-world Arabic database is used to conduct our experimental evaluation where two stemmers and three similarity functions are used to see the effect on DySimII with Arabic dataset. The results represent that matching accuracy is improved using Asem stemmer when the number of corrupted attributes is increased, also testing the three similarity functions show that using winkler similarity function provides better matching accuracy while N-gram provides better results when used with Asem stemmer.

The conventional model predictive control (MPC) suffers from high common-mode voltage (CMV) magnitude and large current ripples. In this article, the reduced CMV MPC strategy with double vector (RCMV-MPCDV) is proposed for three-level inverters, which adopts double vector to reduce the CMV and current ripples simultaneously in per-sampling period. First, based on geometric relationship of median, a novel cost function that evaluates median line distance is proposed to reduce the computational complexity, rather than evaluating the closest distance. Second, in order to reduce the computational burden, the double vector preselection algorithm is presented with fewer evaluation times. Third, the candidate vectors are reclassified and regrouped from partial basic vectors, which restricts CMV magnitude within one-sixth of dc-link voltage. In addition, to select more satisfactory double vectors for RCMV-MPCDV scheme, two improved methods are proposed to compensate the losing candidate vectors. The simulation and experimental results with steady-state performance and dynamic response validate the effectiveness of the proposed method.

Sentence similarity determines whether two sentences are close in their structure and meaning. The detection of sentence similarity can be affected by several factors such as sentence representation, similarity measure, and words weighting function. In this study, the impact of three factors that influence similarity detection and paraphrasing identification is evaluated using clustering algorithms. In the evaluation of the impact of these factors, we tried different word embedding models, clustering algorithms, and weighting methods for the context words. The clustering algorithms are applied to an Arabic paraphrasing benchmark that consists of 1010 pairs of Arabic sentences constructed on the basis of Arabic transformation rules and labeled for similarity and paraphrasing. Experimental results show that pre-trained embedding, weighting context words with part of speech, and labeling sentence pairs by the majority of experts provides better recall and precision.

We introduce a recently developed nonlinear-type time-frequency analysis tool, synchrosqueezing transform (SST), to quantify complicated and noisy physiological waveform that has time-varying amplitude and frequency. We apply it to analyze a peripheral venous pressure (PVP) signal recorded during a seven hours aortic valve replacement procedure. In addition to showing the captured dynamics, we also quantify how accurately we can estimate the instantaneous heart rate from the PVP signal.

Public eagerness to use Transjakarta as a transport mode has been increasing over the past few years as the ridership growth consistently. To keep the growth up, stakeholders need to understand the passenger behavior by creating an effective operational system so that the passenger satisfaction maintained. An understanding of passenger behavior can be obtained by identifying its pattern and trend which recorded as passenger transaction data. This project uses passenger transaction data which contains information about time and space to produce interactive visualization as a means to identify passenger behavior. Interactive visualization created in this project depicts the trend of bus stop occupancy and passenger movement from one stop to another. The visualization consists of a map and diagrams which relate to each other as one integrated analytical dashboard system. The visualization design process combines Tableau 2019.2 and Mapbox GL JS to create powerful visualization supported by a 3D map view that facilitates analytical reasoning work. Geovisual analytics dashboard as a result of this project is intended for PT. Transportasi Jakarta staff to make decisions. Some staff was asked to examine the usability of the dashboard regarding its effectiveness, efficiency, and accessibility aspects. The usability test pointed out that the analytical dashboard can generate new information about the patterns and trends of Transjakarta passengers.

Processor power management exploiting Dynamic Voltage and Frequency Scaling (DVFS) plays a crucial role in improving the data-center’s energy efficiency. However, we observe that current power management policies in Linux (i.e., governors) often considerably increase tail response time (i.e., violate a given Service Level Objective (SLO)) and energy consumption of latency-critical applications. Furthermore, the previously proposed SLO-aware power management policies oversimplify network request processing and ignore the fact that network requests arrive at the application layer in bursts. Considering the complex interplay between the OS and network devices, we propose a power management framework exploiting network packet processing mode transitions in the OS to quickly react to the processing demands from the received network requests. Our proposed power management framework tracks the transitions between polling and interrupt in the network software stack to detect excessive packet processing on the cores and immediately react to the load changes by updating the voltage and frequency (V/F) states. Our experimental results show that our framework does not violate SLO and reduces energy consumption by up to 35.7% and 14.8% compared to Linux governors and state-of-the-art SLO-aware power management techniques, respectively.

The power router (PR) based on power electronic technology is the core equipment of the energy internet (EI). For the PR, the common direct current (DC) bus structure is drawing more and more attention, in which the DC voltage stability is of vital importance. In the PR with the energy storage system (ESS), the DC/DC converter of ESS has significant effects on the DC bus voltage stability. Droop control is commonly used in the control of DC/DC converters in ESS. However, conventional droop control fails to provide the inertia to meet the demand for voltage stability. Hence, a virtual inertia control strategy with load current feedforward is proposed in this paper to strengthen the voltage stability. Firstly, the small-signal model of ESS is derived to find out the key factors that affect the DC bus voltage stability. Secondly, based on the analysis, an inertial droop control strategy is presented to provide inertia to the DC bus voltage, while a load current feedforward method is proposed to improve the voltage response performance during load transients. The stability analysis is also carried out to guide the controller designing. The validity of the proposed control strategy has been verified by the results from both MATLAB/Simulink and hardware-in-the-loop (HIL) simulation.

In order to compensate unbalanced and non-linear loads of the low-voltage distribution network into the symmetric and pure resistive loads, a novel method is proposed, where the compensation current calculation is performed based on the instantaneous voltage space vector orientation. The proposed method is capable of making the waveform and phase of the compensated current consistent with those of the grid voltage, so as to achieve the minimum instantaneous reactive power consumption under the asymmetric and distorted background of the grid. It is found that this method has considerable advantages, including the low computational expense, adaptive phase sequence and the automatic frequency following, over the conventional methods. The accuracy and effectiveness of the proposed method are verified through simulation and experiments. It is concluded that the proposed method is an accurate and effective method for shunt comprehensive compensation in the low voltage distribution network.

Energy router (ER), as the core equipment of energy internet (EI), can realize the regulation, conversion, and storage of multiple energy carriers. The conversion efficiency of ERs is important which is affected by load or temperature varying with actual operating environment. But, the efficiencies of conversion devices are assumed to be constant in the conventional ER model, which results in unreasonable operation schemes. Then, the key factors affecting the economic dispatch of the ER is the accuracy of the model. Therefore, an improved energy hub (EH) dynamic model is proposed for representing complex power flows of the physical layer, where conversion efficiency varies with temperature or load rate. Besides, to further improve the flexibility of the model, energy storage and region exchange matrix are added to the proposed model. Finally, the feasibility and effectiveness of the proposed model are verified based on the general algebraic modeling system (GAMS) in a residential area.

The input-series-output-parallel (ISOP) hybrid dc transformer (HDCT) combining series resonant converter (SRC) and dual-active bridge (DAB) converter has higher efficiency than the ISOP DAB converter while maintaining the controllability of dc voltage or power. However, the startup control of ISOP HDCT is quite challenging, and the incorrect and uncoordinated startup strategies may cause excessive voltage and current stress in SRC and DAB. This paper proposed a three-stage startup strategy to address these issues, which can cooperatively control two types of converter submodules, accurately limit the switching current, quickly eliminate the overvoltage on input capacitors, and set up the output voltage with negligible overshoot. Besides, through the proposed strategy, SRC and DAB can smoothly shift to serial resonant and single phase-shift (SPS) modulation mode, respectively, which will not affect the steady-state control performance. The proposed strategy is experimentally verified on a 500 W prototype.

BACKGROUNDOne of the main weaknesses and problems in the field of road accident management is the lack or non-use of appropriate information systems. The overall aim of this study is to reduce the impact of road accidents in the west, southwest, and south routes of Tehran Province, Iran, by designing Management Database responsible for road accidents.METHODSThis was an applied study performed using the descriptive-analytical method in which the data entry and analysis were conducted in the Geographic Information System (GIS) environment in line with the objectives. Then, according to different requirements of the various users, an appropriate analysis was performed on the layers and the software spatial analysis was applied in line with the main objective of the study. In the next step, the desired outputs were extracted from the system and the main output, i.e. the comprehensive database of spatial data required to manage road accidents, was provided to the management levels of various organizations for planning.FINDINGSIn this study, the accident-prone areas in the studied routes were identified using the kernel density estimation (KDE) method.CONCLUSIONPerforming various analyses on geographical and descriptive information and providing appropriate outputs and also sharing them can play an effective role in reducing injuries caused by road accidents.

Health monitoring of rotating machinery is commonly based on vibration signals. Instead, this pioneering research provides bearing diagnostics using strain measurements, obtained from Fiber Bragg Grating (FBG) fiber-optic sensors. Besides detection of the damage via spectral data, these optical sensors also allow the estimation of damage severity through the direct and accurate measurement of the damage size of small spalls in the bearing races, an essential capability for the prognosis of remaining useful life. FBG sensors are small and can be easily placed in the immediate proximity of the bearing or even embedded inside it, thereby ensuring much enhanced signal-to-noise ratio through the minimizing transmission path effects from remote disturbances. These ball-bearing related diagnostic capabilities of FBG sensors are demonstrated via seeded tests, as well as by means of extended monitoring of bearings during fatigue endurance tests. Sensitivity to FBG sensor location is studied, showing acceptable values at all housing measuring points around the bearing. Fiber-optic sensors appear to have promising diagnostic potential for spall-like faults in both the outer and inner races of ball bearings with a very good discrimination power.

Recently, a few fault-tolerant control strategies have been proposed for T-type three-level inverters to improve the reliability of the inverter system. However, the neutral-point (NP) voltage oscillations suppression is never considered in the previous fault-tolerant control strategies. In order to suppress the NP voltage oscillations when open-circuit fault occurs to NP switches, this paper proposes two novel fault-tolerant control strategies named as improved two-level switching tolerant control and hybrid switching tolerant control (HS-TC). Both methods are implemented by adding an optimal time-offset to the three-phase turn- on times. The corresponding time-offset value of each method is simply calculated considering the phase current and dwell time of [O] state. Moreover, an NP voltage balancing algorithm to eliminate the dc deviation and a dc bias to further reduce the NP voltage oscillations are used in the HS-TC. The performance and comparison of the tolerant controls with respect to the NP voltage oscillations, current total harmonic distortion, and dynamic response are illustrated and analyzed through the simulation and experimental results.

The parallel-operated three-level T-type inverters (3LT2Is) are increasingly used for large capacity renewable energy sources, which can increase power rating, efficiency, and reliability. However, the circulating current problem arises. Circulating current inevitably distorts output currents, increases power losses, and reduces the efficiency. Unbalanced operating conditions, including both unbalanced grid voltage and unbalanced filter parameters, cause adverse effects on the operation of parallel 3LT2Is. In this paper, a hybrid control scheme is developed to suppress the circulating current for parallel 3LT2Is under unbalanced operating conditions. An average model is derived and analyzed. Two quasi-resonant controllers are added to the conventional PI controller, and feed-forward control terms are introduced to eliminate the effects of disturbances in a zero-axis current system. Circulating current suppression is realized by real-time adjusting dwell times of small vectors in space vector modulation technique. A proportional controller is adopted to balance the neutral point voltage. Simulation and experimental results validate theoretical analysis and the effectiveness of the proposed scheme.

The COVID-19 outbreak began in December 2019 and has dreadfully affected our lives since then. More than three million lives have been engulfed by this newest member of the corona virus family. With the emergence of continuously mutating variants of this virus, it is still indispensable to successfully diagnose the virus at early stages. Although the primary technique for the diagnosis is the PCR test, the non-contact methods utilizing the chest radiographs and CT scans are always preferred. Artificial intelligence, in this regard, plays an essential role in the early and accurate detection of COVID-19 using pulmonary images. In this research, a transfer learning technique with fine tuning was utilized for the detection and classification of COVID-19. Four pre-trained models i.e., VGG16, DenseNet-121, ResNet-50, and MobileNet were used. The aforementioned deep neural networks were trained using the dataset (available on Kaggle) of 7232 (COVID-19 and normal) chest X-ray images. An indigenous dataset of 450 chest X-ray images of Pakistani patients was collected and used for testing and prediction purposes. Various important parameters, e.g., recall, specificity, F1-score, precision, loss graphs, and confusion matrices were calculated to validate the accuracy of the models. The achieved accuracies of VGG16, ResNet-50, DenseNet-121, and MobileNet are 83.27%, 92.48%, 96.49%, and 96.48%, respectively. In order to display feature maps that depict the decomposition process of an input image into various filters, a visualization of the intermediate activations is performed. Finally, the Grad-CAM technique was applied to create class-specific heatmap images in order to highlight the features extracted in the X-ray images. Various optimizers were used for error minimization purposes. DenseNet-121 outperformed the other three models in terms of both accuracy and prediction.

Unbalanced grid is a normal fault in power system, and in order to ensure the rectifier current tracking precision under this condition, the design and implementation of rectifier control strategies become more complex. In view of this, a simplified grid current tracking strategy is proposed, which is suitable for LCL-based rectifier operating in both balanced and unbalanced conditions. The grid current sinusoidal is achieved by tracking converter-side current of LCL filter. The current reference is only calculated by grid voltage and its 1/4 period lagging signal at stationary coordinates. Compared with conventional strategies, sequence separation and Phase-Locked Loop (PLL) are avoided. Besides, an improved active damping method is applied, which can accurately extract harmonic voltage and enhance the resonance suppressing effect. Simulation and experimental results verify the effectiveness of the proposed strategy.

DAB-LLC sigma converter has been an attractive topology, which combines the high efficiency of LLC and the high controllability of DAB. However, traditional PI controller manages output voltage/power by adjusting the phase shift angle of DAB and allows LLC to operate in an open loop mode, which leads to poor transient response. In this paper, an improved simplified optimal trajectory controller (SOTC) suitable for LLC in sigma converter is investigated. And then a new hybrid control strategy is proposed, which can not only control DAB by PI controller, but also control LLC by improved SOTC at transient. The hybrid control strategy compensates the load current and enhances the intrinsic voltage feedback of the sigma converter. Furthermore, it reduces current overshoot and improves transient response without reducing the efficiency. Finally, a prototype rated at 1 kW is built and tested, and experimental results verified the effectiveness of proposed strategy.

In modern server CPUs, last-level cache (LLC) is a critical hardware resource that exerts significant influence on the performance of the workloads, and how to manage LLC is a key to the performance isolation and QoS in the cloud with multi-tenancy. In this paper, we argue that in addition to CPU cores, high-speed I/O is also important for LLC management. This is because of an Intel architectural innovation - Data Direct I/O (DDIO) - that directly injects the inbound I/O traffic to (part of) the LLC instead of the main memory. We summarize two problems caused by DDIO and show that (1) the default DDIO configuration may not always achieve optimal performance, (2) DDIO can decrease the performance of non-I/O workloads that share LLC with it by as high as 32%. We then present IAT, the first LLC management mechanism that treats the I/O as the first-class citizen. IAT monitors and analyzes the performance of the core/LLC/DDIO using CPU's hardware performance counters and adaptively adjusts the number of LLC ways for DDIO or the tenants that demand more LLC capacity. In addition, IAT dynamically chooses the tenants that share its LLC resource with DDIO to minimize the performance interference by both the tenants and the I/O. Our experiments with multiple microbenchmarks and real-world applications demonstrate that with minimal overhead, IAT can effectively and stably reduce the performance degradation caused by DDIO.

In this paper, a comprehensive characterization and calibration of a K-band large scale active phased-array antenna (APAA) that consists of 1024 (32×32) radiating elements based on a modular architecture is presented. RF-channels are fully characterized inside the APAA environment using a planar near-field probe setup. A very accurate and efficient calibration technique is suggested and verified successfully over a wide range of scan angles (0 to 70 degree) and wide frequency band of 18-21 GHz.

The process of selecting the appropriate meaning of an ambigous word according to its context is known as word sense disambiguation. In this research, we generate a number of Arabic sense inventories based on an unsupervised approach and different pre-trained embeddings, such as Aravec, Fast text, and Arabic-News embeddings. The resulted inventories from the pre-trained embeddings are evaluated to investigate their efficiency in Arabic word sense disambiguation and sentence similarity. The sense inventories are generated using an unsupervised approach that is based on a graph-based word sense induction algorithm. Results show that the Aravec-Twitter inventory achieves the best accuracy of 0.47 for 50 neighbors and a close accuracy to the Fast text inventory for 200 neighbors while it provides similar accuracy to the Arabic-News inventory for 100neighbors. The experiment of replacing ambiguous words with their sense vectors is tested for sentence similarity using all sense inventories and the results show that using Aravec-Twitter sense inventory provides a better correlation value

In a multi-tenant environment, the companies referred to as tenants share a common application and Relational Database Management System (RDBMS) instances to store their data. However, with the rapid adoption of multi-tenant databases, the cloud provider faces two challenges: Tenants have irregular workload patterns. Also, tenants require a strict guarantee of their rental services’ quality and performance, known as a Service Level Agreement (SLA). In this research, a Multi-Tenant Database Management System (MT DBMS) is presented. A multi-tenant migration algorithm called MT-M is presented, which migrates the violated tenants on an elastic cluster of machines to mitigate the SLA Violations. Experiment results show that the proposed MT-M algorithm is ideal for the migration of the violated multi-tenant databases, reducing SLA violations’ total number compared to the previous migration algorithms.

The Home Energy Management System (HEMS) can reduce the cost of electricity for users and improve the stability of the grid. However, the uncertainty of scheduling inputs is the major challenge for HEMS. Thus, a real time energy management strategy considering the uncertainty is proposed in this paper. The proposed method includes two parts: the improved day-ahead energy management (DAEM) and a novel real time energy management (RTEM). For DAEM, considering the needs for charging and discharging the next day, this paper improves the DAEM. Then based on the improved DAEM, a novel RTEM is proposed to cope with the uncertainty of electricity price and renewable energy forecasting. Guided by the SOC results of improved DAEM, the RETM schedules for each hour are adjusted based on updated schedule variable inputs. The simulation results show that proposed strategy can improve the performance of household electricity dispatching and reduce the impact of uncertainty on the system.

There are inevitably unbalanced states in modular multilevel converter (MMC), including switching states imbalance and parameters asymmetry in practical engineering, which will lead to energy imbalance, power loss, device damage and reliability decrease. To solve these issues, an improved model predictive control (MPC) based energy balance method is proposed in this paper. In the proposed method, the capacitor energy is analyzed and as the control objective directly, which aviods complicated asymmetric model analysis for MMC. Firstly, the relationship between the circulating current and arm energy is analyzed in detail. Then, the unbalanced components of the capacitor voltage of sub-modules (SMs) is injected into circulating current reference to control the energy between and within phases in balance. The proposed method achieves energy balance without influencing circulating current and increasing computational complexity. The feasibility and validity of the proposed control strategy are demonstrated by simulation results using Matlab/Simulink.

The stable and quick restoration is necessary for resilient microgrids. This paper proposes the new concept of grey start and its strategy as a restoration mode for hybrid AC/DC microgrid in semi-black state when the system blackout occurs. Firstly, an optimal reconstruction model of the whole network is established and solved by biogeography-based optimization (BBO) algorithm. Secondly, the classification and evaluation model of distributed generations (DGs) is established by using improved BP artificial neural network (ANN). Thirdly, optimal path planning is realized. In addition, to improve the stability during the process of grey start restoration, the control methods of different DGs after classification are discussed. In comparison with conventional black start restoration mode, the restoration time is shortened, and the fluctuation of bus voltage is suppressed. The feasibility of the proposed grey start strategy is verified by the results of numerical example analysis and MATLAB/Simulink simulation, respectively.

Word sense disambiguation is the process of determining the proper meaning of a word according to its context. In this study, we represent the impact of word embedding on building Arabic sense inventory by an unsupervised approach. Three pre-trained embeddings are tested to investigate their effect on the resulting sense inventory and their efficiency in word sense disambiguation for Arabic context. Sense inventories are constructed using a fully unsupervised method based on graph-based word sense induction algorithm. The results show that Aravec-Twitter inventory achieves the best accuracy of 0.47 for 50-neighbors and a close accuracy to the Fasttext inventory for 200-neighbors.

In this work, we will address the opportunities of a hybrid III-V/CMOS technology for next generation wireless communication, beyond 5G, moving to operating frequencies above 100GHz. Challenges related to III-V upscaling and CMOS co-integration using 3D technologies will be discussed.

Photoplethysmography (PPG) is increasingly used in digital health, exceptionally in smartwatches. The PPG signal contains valuable information about heart activity, and there is lots of research interest in its means and analysis for cardiovascular diseases. Unfortunately, to our knowledge, there is no arrhythmic PPG dataset publicly available-this paper attempt to provide a toolbox that can generate synthesized arrhythmic PPG signals. The model of a single PPG pulse in this toolbox utilizes two combined Gaussian functions. This toolbox supports synthesizing PPG waveform with regular heartbeats and three irregular heartbeats: compensation, interpolation, and reset. The user can generate a large amount of PPG data with a certain irregularity, with different sampling frequency, time length, and a range of noise types (Gaussian noise and multi-frequency noise) can be added to the synthesized PPG which can all be modified from the interface, and different types of arrhythmic PPGs (as calculated by the model) generated. The generation for large PPG datasets that simulate PPG collected from real humans could be used for testing the robustness of developed algorithms that are targeting arrhythmic PPG signals. Our PPG synthesis tool is publicly available.

Deep Neural Networks (DNNs) have reinvigorated real-world applications that rely on learning patterns of data and are permeating into different industries and markets. Cloud infrastructure and accelerators that offer INFerence-as-a-Service (INFaaS) have become the enabler of this rather quick and invasive shift in the industry. To that end, mostly accelerator-based INFaaS (Google’s TPU [1], NVIDIA T4 [2], Microsoft Brainwave [3], etc.) has become the backbone of many real-life applications. However, as the demand for such services grows, merely scaling-out the number of accelerators is not economically cost-effective. Although multi-tenancy has propelled datacenter scalability, it has not been a primary factor in designing DNN accelerators due to the arms race for higher speed and efficiency. This paper sets out to explore this timely requirement of multi-tenancy through a new dimension: dynamic architecture fission. To that end, we define Planaria1 that can dynamically fission (break) into multiple smaller yet full-fledged DNN engines at runtime. This microarchitectural capability enables spatially co-locating multiple DNN inference services on the same hardware, offering simultaneous multi-tenant DNN acceleration. To realize this dynamic reconfigurability, we first devise breakable omni-directional systolic arrays for DNN acceleration that allows omni-directional flow of data. Second, it uses this capability and a unique organization of on-chip memory, interconnection, and compute resources to enable fission in systolic array based DNN accelerators. Architecture fission and its associated flexibility enables an extra degree of freedom for task scheduling, that even allows breaking the accelerator with regard to the server load, DNN topology, and task priority. As such, it can simultaneously co-locate DNNs to enhance utilization, throughput, QoS, and fairness. We compare the proposed design to PREMA [4], a recent effort that offers multi-tenancy by time-multiplexing the DNN accelerator across multiple tasks. We use the same frequency, the same amount of compute and memory resources for both accelerators. The results show significant benefits with (soft, medium, hard) QoS requirements, in throughput (7.4×, 7.2×, 12.2×), SLA satisfaction rate (45%, 15%, 16%), and fairness (2.1×, 2.3×, 1.9×).

Phase is the most fundamental physical quantity when we study an oscillatory time series. There are many tools aiming to estimate phase, most of them are developed based on the analytic function model. Unfortunately, this approach might not be suitable for modern signals with intrinsic nonstartionary structure, including multiple oscillatory components, each with time-varying frequency, amplitude, and non-sinusoidal oscillation, e.g., biomedical signals. Specifically, due to the lack of consensus of model and algorithm, phases estimated from signals simultaneously recorded from different sensors for the same physiological system from the same subject might be different. This fact might challenge reproducibility, communication, and scientific interpretation and thus we need a standardized approach with theoretical support over a unified model. In this paper, after summarizing existing models for phase and discussing the main challenge caused by the above-mentioned intrinsic nonstartionary structure, we introduce the adaptive non-harmonic model (ANHM), provide a definition of phase called fundamental phase, which is a vector-valued function describing the dynamics of all oscillatory components in the signal, and suggest a time-varying bandpass filter (tvBPF) scheme based on time-frequency analysis tools to estimate the fundamental phase. The proposed approach is validated with a simulated database and a real-world database with experts’ labels, and it is applied to two real-world databases, each of which has biomedical signals recorded from different sensors, to show how to standardize the definition of phase in the real-world experimental environment. Specifically, we report that the phase describing a physiological system, if properly modeled and extracted, is immune to the selected sensor for that system, while other approaches might fail. In conclusion, the proposed approach resolves the above-mentioned scientific challenge. We expect its scientific impact on a broad range of applications.

In this paper, a novel method based on a convolutional neural network is presented for receiver phased array online calibration using power-only measurements. Phase calibration coefficient for each RF path of a phased array receiver is estimated by combining a convolutional and multi-layer perceptron neural network in different signal to noise ratio (SNR) conditions. To validate the proposed method, a $1\times 8$ linear active phased array receiver is designed and simulated based on real characteristics. Compared to other conventional techniques such as the rotating element electric field vector (REV) method, our proposed method requires less number of power measurements and hence, is less time-consuming. Moreover, the proposed method has a significantly better performance in the presence of noisy measurements. Simulation results show that a mean-squared-error (MSE) of 6.5 is achieved at the signal to noise ratio of 20 dB for one iteration of the power measurement. These results prove that the proposed method can be adopted to the phased array online calibration, to have a faster calibration with higher accuracy in noisy conditions in comparison with conventional methods.

I/O performance plays a critical role in the overall performance of modern servers. The emergence of ultra high-speed I/O devices makes the data movement between processors, main memory, and devices a major performance bottleneck. Conventionally, the main memory is used as an intermediate buffer between the processor and I/O devices and I/O devices cannot directly access processor side caches. Data Direct I/O (DDIO) technology aims to reduce the memory bandwidth utilization by enabling the I/O devices to leverage Last Level Cache (LLC) as the intermediate buffer. Our experimental results show that DDIO can completely eliminate memory bandwidth utilization while running network-or storage-intensive applications. However, when modeling the I/O subsystem using architectural simulators, DDIO is often ignored, which can result in inaccurate assessments about the I/O and memory sub system of emerging and future large-scale computer systems. In this paper, we provide a detailed background on DDIO technology in Intel server processors. Then we present our cycle-accurate I/O subsystem model in gem5 simulator that can be configured to model DDIO. We verify our model against baseline gem5 and validate it by comparing its results against a physical computer system.

Although social media and contents are being generated and shared with an unprecedented scale and speed, rural and underdeveloped areas throughout the world have only limited access due to the lack of high-speed Internet. Connecting rural communities to the digital world and providing them with right contents will provide the much-needed bridge between urban and rural areas. In this paper, we propose a communication and information framework that utilizes simple Internet of Things (IoT) in a rural community to assist delay-tolerant content distribution. Specifically, a hybrid fog-cloud content distribution network is constructed by deploying low-end simple fog nodes and utilizing the movement of community vehicles. Moreover, drones are used to distribute content on demand as a compliment of the delay tolerant network for better delivery rate and lower delay time. A novel drone scheduling algorithm is proposed to plan drones’ tours optimally. Extensive simulation experiments have been performed to evaluate the performance of the proposed framework.

Rewriting sentences using different words leading to the same meaning of the original sentence is called paraphrasing while paraphrasing identification task is the task of detecting the sentence that is paraphrased from another. This research provides a literature survey of researches that have studied and proposed methods for paraphrasing identification of English and Arabic texts. Accordingly, the impact of paraphrasing on NLP applications and paraphrasing benchmarks are also studied. The comparative study on the available paraphrasing identification techniques shows that the best precision is provided with WordNet based techniques while the best accuracy is provided by deep learning with statistical features of English texts. In the case of Arabic, the best precision is provided by the distributed word vector representations with Convolutional Neural Network.

GaN-channel based transistors are ideally suited for RF/5G applications and also provide the promise of monolithic integration on a conventional Si-platform. Due to the wide scope of high electron mobility GaN transistor architectures, their reliability assessment is essential to ensure their successful deployment in low-power applications such as mobile computing devices, as well as high-power applications such as autonomous vehicles and base stations. We identify the most important DCreliability metrics necessary for fair benchmarking of future GaNon-Si RF transistors. A detailed analysis of the shortlisted DCreliability parameters for three device types, namely MOSFETs, MOSHEMTs/MISHEMTs and HEMTs is presented. MOSHEMT/MISHEMT is identified as the most robust device architecture, due to the presence of a barrier layer alleviating the impact of certain degradation mechanisms. Defect distributions in the gate-stack of MOS devices are extracted using defect band modelling technique. MOSHEMT devices are shown to undergo negative and positive Bias Temperature Instability (BTI) under specific ranges of positive gate-overdrive, thereby demonstrating the importance of correctly estimating the oxide field for MOSHEMT devices. Degradation map methodology is partially developed to distinguish the different gate-oxide degradation mechanisms and model the device lifetime pertaining to each of the mechanisms.

Multilevel voltage source inverters are superior solutions for medium-voltage (MV) and high power applications, which can meet the requirements of high power density, efficiency and power quality. This paper presents a novel Neutral-Point-Clamped Half-Bridge (NPC-HB) eleven-level inverter topology which is composed of two power switches at fundamental frequency connected in series with the positive and negative poles of the DC voltage source and a seven-level NPC-HB inverter. Only ten switching devices are needed to generate eleven voltage levels, which improves the power density of the inverter system with fewer switches. Besides, the DC voltage utilization ratio of the proposed topology is higher than 1. After the comprehensive analysis of the operation principles of the proposed topology, a proposed compound modulation strategy based on phase-disposition pulse width modulation (PD-PWM) suitable for the eleven-level inverter is described in detail. With the proposed modulation strategy, the violent fluctuation of flying capacitor voltage is solved when the modulation degree is high. Simulation and experimental results verify the validity of the proposed topology and modulation strategy.

With the integration of distributed generation systems, power electronic equipments, and non-linear loads, there are more risks of power quality disturbances (PQDs) in the microgrid. In order to ensure reliable and quality supply of power, the automatic classification of power quality is very important. In this paper, a novel deep learning method based on deep convolution neural network (CNN) and spectrogram for PQDs classification is proposed. Considering the characteristics of PQDs problems, the spectrogram is used to restructure PQDs waveforms, and the convolutional base, which consists of 2-D convolution, batch-normalization, dropout and max pooling layers, is designed to capture features, speed up training, and reduce overfitting. Through the convolutional base, the network can automatically extract features from waveforms of different time length. Moreover, a novel tagging method for PQDs is proposed and the small sample training is realized. The performance of the proposed method is evaluated on a set of synthetic single and combined PQDs waveforms.

Paralleled three-level T-type inverters (3LT2Is) have the merits of high power rating, improved reliability and convenient maintenance. However, there exists circulating current which will lead to current distortion and increased system losses. Especially in unbalanced operating conditions, the circulating current is even more serious, and the conventional control methods are difficult to reduce the circulating currents. To overcome this drawback, a generalized model for analyzing circulating currents in parallel 3LT2Is under unbalanced conditions including unequal grid voltage amplitude and inductance imbalance is developed, and a novel deadbeat control scheme is proposed for the circulating current suppression with the advantages of easy to implement and no inter-module communication. Compared to the conventional methods, controller proposed by this paper is simpler in design and parallel expansion is easier while the influence of unbalanced operating conditions is eliminated. The effectiveness of the proposed method is verified by simulation and experiment results.

This paper proposes a zero-sequence circulating current (ZSCC) control scheme for the multiparalleled three-level T-type inverters (3LT2Is). Based on the analysis with the proposed generalized model of multiparalleled 3LT2Is, it is seen that the ZSCC is influenced by the circuit parameters, duty cycles, and the circulating currents of the l th and first 3LT2I. To mitigate the circulating currents, a deadbeat control strategy for n-paralleled (n≥3) inverters is proposed, which is realized by adjusting the duty cycle of the small vectors in each PWM sampling period. Compared with the conventional PI control method, the ZSCC spike is mitigated with the proposed scheme, thus improving the quality of the output currents. Also, the reliability of the 3LT2Is system is improved because such a scheme only needs the information about the ZSCC and zero-sequence duty ratio in the l th and the first inverter. The proposed method can also be applied to the neutral-point-clamped converters. Simulation and experimental results verify the effectiveness of the theoretical analysis and the proposed method.

Many modern embedded CPUs adopt Non-Coherent DMA (NC-DMA) over Coherent DMA (C-DMA) because of simplicity. An NC-DMA design, however, requires a CPU device driver to explicitly invalidate or flush a wide range of cache space. When an I/O DMA device writes data to a main memory region, the CPU needs to invalidate the cache space corresponding to the same memory region twice: (1) to prevent dirty cache lines from overwriting the DMA data and (2) to remove any cache lines prefetched before the DMA is done. In this work, we first show that such explicit invalidations consume 31 percent of CPU cycles, limiting the data transfer throughput of a high-speed network interface card (NIC) when receiving network packets. Second, we propose a Semi-Coherent DMA (SC-DMA) architecture for improving the efficiency of NC-DMA with a slight modification to the hardware. Specifically, our SC-DMA records the DMA region and prohibits any data that is prefetched from the region from entering the cache, reducing nearly 50 percent of the unnecessary invalidations. Lastly, we identify several software optimizations that can substantially reduce excessive cache invalidations prevalent in NIC drivers. Our evaluation with NVIDIA Jetson TX2 shows that our proposed SC-DMA design with the NIC driver optimizations can improve the NIC data transfer throughput by up to 53.3 percent.

Attribute values in textual datasets are subjects of different types of errors due to the data entry processes such as typographical errors, pronunciation errors or dialects alterations. These errors make the entity resolution process more challenging. The iterative blocking indexing technique can be used for correcting this type of errors mainly in query access where the records are stored into more than one block. Blocking indexing technique selects a subset of object pairs saved in the same block for later detailed computation for similarity discarding other pairs in other blocks considering them as irrelevant. This work aims to solving such problems for Arabic texts. It proposes to adapt a specific model for learning blocking keys and analyze its performance for Arabic datasets. The resulted blocking keys are passed as blocking keys to Dynamic Aware Inverted Index (DySimII) that worked efficiently with Arabic datasets. The model is tested against a telephone book dataset that contains duplicates and errors in attribute values according to phonetic and typing errors. The results reach a matching accuracy of 84% for using learned keys with small number of corrupted attributes while the performance is declined with the increase of the number of corrupted attributes.

As the penetration of renewable power generation units connected to the grid increases, high power quality and high efficiency have been widespread concerned. The parallel 3LT 2 Is are necessary to improve the quality and efficiency of applications. The use of common dc-link without isolating transformers will cause the zero-sequence circulating current (ZSCC) problem. This study proposes a finite control set model predictive current control to suppress the ZSCC, balance the neutral-point (NP) voltage as well as control the currents with fast transient response. The cause of the ZSCC generated in the parallel 3LT 2 I is analysed. The proposed ZSCC suppression technique is achieved through changing the redundant small vectors in the parallel 3LT 2 Is. Moreover, the coupling problem between the ZSCC and NP balance is solved by the amplitude comparison of NP voltage and ZSCC. As a result, better ZSCC suppression and NP voltage performance can be achieved in different output currents and filter inductance condition. The proposed method has been verified experimentally.

Indonesia is one of the nations located in the Ring of Fire. Indonesia has a high level of geodynamic activities so that it's often earthquake tectonics. The earthquakes are caused by Indonesia position located in the confluence of four main plates. At present, the history of earthquake data in Indonesia has been accessible by the public. However, general visualization which can present history earthquake in the form maps and summary statistics have not been available. Therefore, this research aims to visualize the history of earthquake data interactively combining spatial data and temporal data. The data used for this research was obtained from BMKG website. The data variables used in this research include CORS stations and history of earthquake phenomenons between 2004 and 2019. The earthquake phenomenon consists of occurrence time, coordinate position, depth and magnitude. The data are processed using Ms Excel and ArcGIS Online Map then are visualized by Web AppBuilder for ArcGIS. The results of the data processing are maps presented in a dashboard with time-series animation and widgets features. We performed maps, graphics and time-series animation as interactive visual interfaces and matched the tasks to visual analytics techniques that are capable to support them. In this paper, we introduce the relationship between variables and present the visual analytics techniques using several example scenarios of Spatio-temporal earthquake data.