Find Research Output

Research Output
  • All
  • Scholar Profiles
  • Research Units
  • Research Output
Filter
Department Publication Year Content Type Data Sources

SELECTED FILTERS

Department of Electrical and Electronic Engineering
Clear all

1.Moving shadow detection via binocular vision and colour clustering

Author:Lu, L;Xu, M;Smith, JS;Yan, YY

Source:IET COMPUTER VISION,2020,Vol.14

Abstract:A pedestrian segmentation algorithm in the presence of cast shadows is presented in this study. The novelty of this algorithm lies in the fusion of multi-view and multi-plane homographic projections of foregrounds and the use of the fused data to guide colour clustering. This brings about an advantage over the existing binocular algorithms in that it can remove cast shadows while keeping pedestrians' body parts, which occlude shadows. Phantom detection, which is inherent with the binocular method, is also investigated. Experimental results with real-world videos have demonstrated the efficiency of this algorithm.

2.AlGaN/GaN Metal-Insulator-Semiconductor (MIS)-HFETs Based DC-DC Boost Converters with Integrated Gate Drivers

Author:Cui, M;Bu, QL;Cai, YT;Sun, RZ;Liu, W;Wen, HQ;Lam, S;Liang, YC;Mitrovic, IZ;Taylor, S;Chalker, PR;Zhao, CZ

Source:2019 10TH INTERNATIONAL CONFERENCE ON POWER ELECTRONICS AND ECCE ASIA (ICPE 2019 - ECCE ASIA),2019,Vol.

Abstract:This study proposed a 100 kHz, 5V/11V boost converter with an integrated gate driver for a power switching device using recessed E-mode MIS-HFETs. The integrated gate driver consisting of multi-stages DCFL (Direct-Coupled FET Logic) inverters and a buffer stage, has large input swing (up to 10 V) and wide noise margin with gate dielectric, which benefits applications requiring large gate swing without any additional drivers or level shifters. The impact of transistor size on rise times and fall times have been studied. Either buffer stage or larger width of DCFL inverter can reduce rise times from 2.4 mu s to less than 0.5 mu s at 100 kHz, so the output voltage of boost converter is increased by 10 %% at a duty cycle of 0.7. However, large buffer width can result in high gate overshoot and oscillation, indicating careful design to balance switching speed and oscillation.

3.Advances of RRAM Devices: Resistive Switching Mechanisms, Materials and Bionic Synaptic Application

Author:Shen, ZJ;Zhao, C;Qi, YF;Xu, WY;Liu, YN;Mitrovic, IZ;Yang, L;Zhao, CZ

Source:NANOMATERIALS,2020,Vol.10

Abstract:Resistive random access memory (RRAM) devices are receiving increasing extensive attention due to their enhanced properties such as fast operation speed, simple device structure, low power consumption, good scalability potential and so on, and are currently considered to be one of the next-generation alternatives to traditional memory. In this review, an overview of RRAM devices is demonstrated in terms of thin film materials investigation on electrode and function layer, switching mechanisms and artificial intelligence applications. Compared with the well-developed application of inorganic thin film materials (oxides, solid electrolyte and two-dimensional (2D) materials) in RRAM devices, organic thin film materials (biological and polymer materials) application is considered to be the candidate with significant potential. The performance of RRAM devices is closely related to the investigation of switching mechanisms in this review, including thermal-chemical mechanism (TCM), valance change mechanism (VCM) and electrochemical metallization (ECM). Finally, the bionic synaptic application of RRAM devices is under intensive consideration, its main characteristics such as potentiation/depression response, short-/long-term plasticity (STP/LTP), transition from short-term memory to long-term memory (STM to LTM) and spike-time-dependent plasticity (STDP) reveal the great potential of RRAM devices in the field of neuromorphic application.

4.IEEE Access Special Section: Emerging Technologies for Energy Internet

Author:Wen, HQ;Liang, YC;Mitrovic, IZ;Li, DP;Tayahi, M;Lu, F;Ye, XM

Source:IEEE ACCESS,2020,Vol.8

Abstract:Renewable energy sources such as photovoltaic (PV), wind, tidal, and ocean waves have increasingly penetrated the global production of energy. Energy Internet has been widely regarded as one of the promising solutions for the serious energy crisis and environmental pollution problem. Unlike the conventional centralized power generation structure, energy Internet widely utilizes different types of distributed generations (DGs), which are located closer to the user and generate electric power within distributed networks. Modular energy storage devices (ESDs) such as batteries for electric vehicles can effectively complement the function of DGs through bidirectional plug-and-play power interfaces. Thus, energy Internet has the ability to minimize power loss, enhance power quality, and improve system reliability.

5.Multimodal Optimization Using Particle Swarm Optimization Algorithms: CEC 2015 Competition on Single Objective Multi-Niche Optimization

Author:Cheng, S;Qin, QD;Wu, Z;Shi, YH;Zhang, QY

Source:2015 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC),2015,Vol.

Abstract:The aim of multimodal optimization is to locate multiple peaks/optima in a single run and to maintain these found optima until the end of a run. The results of seven variants of particle swarm optimization (PSO) algorithms on IEEE Congress on Evolutionary Computation (CEC) 2015 single objective multi-niche optimization problems are reported in this paper. The PSO algorithms include PSO with star structure, PSO with ring structure, PSO with four clusters structure, PSO with Von Neumann structure, social-only PSO with star structure, social-only PSO with ring structure, and cognition-only PSO. The experimental tests are conducted on fifteen benchmark functions. Based on the experimental results, the conclusions could be made that the PSO with ring structure performs better than the other PSO variants on multimodal optimization. To obtain good performance on the multimodal optimization problems, an algorithm needs to converge the candidate solutions to the global optima while keep the population diversity during whole search process.

6.Piecewise Linear Approximation Minimum Current Trajectory for Dual Active Full-bridge Bidirectional DC-DC Converter

Author:Luo, NY;Wen, HQ;Bu, QL

Source:2020 IEEE 9TH INTERNATIONAL POWER ELECTRONICS AND MOTION CONTROL CONFERENCE (IPEMC2020-ECCE ASIA),2020,Vol.

Abstract:This paper will introduce a controller based on the piecewise linear approximation minimum current trajectory(PWL-MCT) for the series resonant dual active-bridge(SR-DAB) dc-dc converters based on the minimum current trajectory(MCT). The linear approximation is based on the minimum current trajectory, thus it can maintain an approximately optimized switching loss at full working range. The power flow through the Dual active-bridge DC-DC converter is controlled by adjusting the duty factor for both primary side and secondary side of the converter and the phase shifting angle between each side s voltage signal simultaneously. While applying the proposed controller, the tank rms current optimizing effect of this converter was proved to perform closely enough to the one applied with original minimum current trajectory controller.

7.Deep Learning Based Multistep Solar Forecasting for PV Ramp-Rate Control Using Sky Images

Author:Wen, HR;Du, Y;Chen, XY;Lim, E;Wen, HQ;Jiang, L;Xiang, W

Source:IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS,2021,Vol.17

Abstract:Solar forecasting is one of the most promising approaches to address the intermittent photovoltaic (PV) power generation by providing predictions before upcoming ramp events. In this article, a novel multistep forecasting (MSF) scheme is proposed for PV power ramp-rate control (PRRC). This method utilizes an ensemble of deep ConvNets without additional time series models (e.g., recurrent neural network (RNN) or long short-term memory) and exogenous variables, thus more suitable for industrial applications. The MSF strategy can make multiple predictions in comparison with a single forecasting point produced by a conventional method while maintaining the same high temporal resolution. Besides, stacked sky images that integrate temporalx2013;spatial information of cloud motions are used to further improve the forecasting performance. The results demonstrate a favorable forecasting accuracy in comparison to the existing forecasting models with the highest skill score of 17.7x0025;. In the PRRC application, the MSF-based PRRC can detect more ramp-rates violations with a higher control rate of 98.9x0025; compared with the conventional forecasting-based control. Thus, the PV generation can be effectively smoothed with less energy curtailment on both clear and cloudy days using the proposed approach.

8.Compliance Current Effect on Switching Behavior of Hafnium Oxide based RRAM

Author:Qi, YF;Zhao, C;Fang, YX;Lu, QF;Liu, CG;Yang, L;Zhao, CZ

Source:2017 IEEE 24TH INTERNATIONAL SYMPOSIUM ON THE PHYSICAL AND FAILURE ANALYSIS OF INTEGRATED CIRCUITS (IPFA),2017,Vol.2017-July

Abstract:In this study, we compared the basic switching behaviors of HfO2, Al2O3 and HfAlOx (Hf:Al=9:1) based RRAM with Ti top electrode by setting various compliance currents (1mA, 5mA, 10mA, 15mA). The resistance ratio of HfO2 based RRAM (20 -> 320) increases with compliance current whereas it drops not obviously for Al2O3 based RRAM (85 -> 54). HfAlOx (Hf:Al=9:1)) based one has the best resistance ratio (300-440) and resistance stability. All low resistance state (LRS) resistance values of three samples are around 100 Omega with large compliance current while there is a difference in HRS resistance which causes the ratio difference accordingly. The dominant mechanism of resistive switching is the formation and rupture of the conductive filament composed of oxygen vacancies. The appropriate compliance current selection and doping technology to high-k materials should be considered in further study.

9.Neuromorphic Properties of Memristor towards Artificial Intelligence

Author:Zhao, C;Shen, ZJ;Zhou, GY;Zhao, CZ;Yang, L;Man, KL;Lim, EG

Source:2018 INTERNATIONAL SOC DESIGN CONFERENCE (ISOCC),2018,Vol.

Abstract:Recent implementations of memristors have opened up the possibility of making brain-like artificial intelligence neuromorphic computing systems, including highly scalable and low-power neural networks. In fact, it has been demonstrated that a memristors can be implemented as an artificial synapse or as a protruding core of an artificial neuron. This paper reviews the neuromorphic properties of memristors, as well as the similarities of neural computation, synapses, and neurons.

10.A full GaN-Integrated Sawtooth Generator based on Enhancement-mode AlGaN/GaN MIS-HEMT for GaN Power Converters

Author:Li, XT;Cui, M;Liu, W

Source:17TH IEEE INTERNATIONAL CONFERENCE ON IC DESIGN AND TECHNOLOGY (ICICDT 2019),2019,Vol.

Abstract:AlGaN/GaN metal-insulator-semiconductor high electron mobility transistors (MIS-HEMTs) have superior advantages like low leakage current and large gate swing. To fully use those advantages, this paper proposes a monolithic sawtooth generator circuit, which can generate a 100 kHz sawtooth waveform with a peak-to-peak voltage of around 3.5 V under 10 V power supply. The integrated circuit is calibrated and simulated by Advanced Design System (ADS). Good agreement between simulations and experimental results indicates the feasibility of GaN MIS-HEMTs on high power electronics application.

11.Brain Storm Optimization Algorithms for Optimal Coverage of Wireless Sensor Networks

Author:Wei, M;Shi, YH

Source:2015 CONFERENCE ON TECHNOLOGIES AND APPLICATIONS OF ARTIFICIAL INTELLIGENCE (TAAI),2015,Vol.

Abstract:Optimal coverage plays an essential role in the service quality of wireless sensor networks because the monitoring can be accurate and meaningful if and only if information from all interesting areas is collected. To ensure high coverage and reduce cost in coverage problems, efficient deployments of sensor nodes in wireless sensor networks become the target of coverage optimization. This paper focuses on using the fewest number of sensor nodes to cover more areas in regular or irregular interesting areas with irregular obstacles inside. To optimize the deployment of sensor nodes, in this paper, a brain storm optimization algorithm is utilized and simulation results show that the algorithm performs well on optimizing the coverage percentage and minimizing the needed sensor nodes under complex environments. In addition, the balance of coverage percentage and needed sensor nodes' number can be adjusted according to specific requirements of different networks. For better optimization results in coverage problems, the step size in the generation process of the brain storm optimization algorithm has also been modified to reach higher coverage using fewer sensor nodes under the same environment.

12.Device-to-Device Communications in LTE-Unlicensed Heterogeneous Network

Author:Yuan, H;Guo, WS;Wang, SY

Source:2016 IEEE 17TH INTERNATIONAL WORKSHOP ON SIGNAL PROCESSING ADVANCES IN WIRELESS COMMUNICATIONS (SPAWC),2016,Vol.2016-August

Abstract:In this paper, the authors examine how the envisaged Device-to-Device (D2D) networks can efficiently scale its capacity by utilizing the unlicensed spectrum with appropriately designed LTE-Unlicensed (LTE-U) protocols. The LTE-U Listen Before Talk (LBT) algorithm is adapted for collision avoidance between traditional unlicensed user equipment (UEs), e.g. Wi-Fi UEs, and the LTE-U enabled D2D UEs. By considering different traffic loads, the analysis found that whilst the D2D UEs reduce the unlicensed network capacity, it increases the combined licensed and unlicensed network capacity by 63%%.

13.Analysis of Wind Profile for Wind Power Station Development

Author:Perera,H. D.Milan Ravinath;Wen,Huiqing

Source:2020 IEEE/IAS Industrial and Commercial Power System Asia, I and CPS Asia 2020,2020,Vol.

Abstract:This research is making an approach to identify the wind speed and wind direction measurement analysis, which is one of the key features in designing a wind station. The wind data will give a complete idea about the wind potential of the site, which is the fuel for turbines to generate its maximum power under maximum efficiency. This will discuss about how to use the collected wind data to filter and edit in WAsP climate analysis tool in order to maximize the purity of the analysis and finally will model a wind profile by using actual wind measurements collected for 4 years and which can be used to model a wind power station. Finally, it will create the Observed wind climate (OWC). The wind rose and Weibull will be distributed and fitted in the simulations accordingly. Therefore, these data can be a green indication to proceed for a wind turbine site development.

14.An Adaptive Constant Power Generation Control Scheme with Simple MPP Estimation for Photovoltaic Systems

Author:Zhu, YX;Wen, HQ;Chu, GY;Li, XS

Source:2019 10TH INTERNATIONAL CONFERENCE ON POWER ELECTRONICS AND ECCE ASIA (ICPE 2019 - ECCE ASIA),2019,Vol.

Abstract:With consideration of instability in environmental conditions, the increasing installation of maximum power point (MPP) operational Photovoltaic systems (PVSs) injects an unstable power generation to the grid (such as, overloading) Aiming to reduce the instable generation during continuous peak power generation, an adaptive constant power generation (CPG) algorithm is proposed in this paper. With the proposed control, the system can operate at MPP tracking (MPPT) mode or CPG mode with the detection from system operators. When the operation mode is detected, the control will operate with self-judgment including regulate the initial mode during CPG with a simple MPP initial MPP estimation. With the estimation of MPP and internal step-size regulation, the response speed, an important factor in variable environments, can be improved. In order to evaluate the proposed scheme, simulation and experiment in different condition, such as changing irradiation and changing limit power level are carried out.

15.Resistive Switching Behavior of Solution-Processed AlOx based RRAM with Ni and TiN Top Electrode at Low Annealing Temperatures

Author:Shen,Zongjie;Zhao,Cezhou;Yang,Li;Zhao,Chun

Source:Proceedings - 2019 International SoC Design Conference, ISOCC 2019,2019,Vol.

Abstract:Solution-processed AlO thin film deposited under different annealing temperatures are used to develop metal/AlOx/Pt RRAM devices, with Ni and TiN as the top electrode (TE) to investigate the influence of metal electrode on device performances. In this work, RRAM devices with various performances exhibit typical bipolar resistive switching (RS) characteristics. The difference of work function between the TE and bottom electrode (BE) metals is considered to play a primary role in operation process. With smaller difference of work function, the devices indicate less power consumption and more stable on/off ratio for SET and RESET operations. The Ni/AlOx/Pt devices demonstrate more stable performance with lower SET and RESET operation voltages (10 ), longer retention time ( 10 s) and better endurance( 100 cycle). x 3 4

16.PSO-based Current Stress Optimization for Three-Level Dual Active Bridge DC-DC Converters

Author:Wang, Y;Zhu, YX;Wen, HQ

Source:2020 CHINESE AUTOMATION CONGRESS (CAC 2020),2020,Vol.

Abstract:Dual-active-bridge (DAB) DC-DC converter is widely adopted and applied fur renewable systems as the power interface considering the fluctuation of the output voltage and power from distributed generators. In order to adapt to high voltage and high power application, three-level diode neutral-point-clamed (3L-DNPC) DAB DC-DC converter has been intensively discussed considering its small electromagnetic interference and wide voltage endurance range. Aiming to improve the efficiency of this 3L-DNPC based DAB converter under all conditions, this paper shows an optimization method to reduce the switch current stress under dual-phase shift modulation strategy. Based on particle swarm optimization (PSO), the proposed control becomes simplified and effectiveness for a wide range. Main simulation and experimental results are provided in order to validate the outstanding performance of the proposed optimization method.

17.Control and Protection of DC Microgird with Battery Energy Storage System

Author:Wen, HQ;Zhu, WQ

Source:2016 IEEE INTERNATIONAL CONFERENCE ON POWER ELECTRONICS, DRIVES AND ENERGY SYSTEMS (PEDES),2016,Vol.2016-January

Abstract:Nowadays, renewable energy technologies, such as solar power and wind power, are trend to replace conventional energy resources. In the meantime, DC Microgrid is becoming popular in distribution systems due to the better compatibility and simpler system control strategy. This paper presents a hierarchical coordinated control strategy for DC Microgrid with wind turbine and photovoltaic array under various operating conditions to ensure reliable operation and provide voltage regulation via simulating equivalent model of DC Microgrid in PSCAD. The contents of this paper can be separated into two parts, the configuration of DC Microgrid and the control strategy. The configuration consists four components, AC grid, power generation unit, power consumption unit and battery energy storage system. The basic principle of those components will be introduced in detail in this paper. The proposal control strategy contains three control level for different objectives. The DC Microgrid model is demonstrated by using PSCAD/ETMDC and the simulation results are also presented to show the effectiveness of the control strategy.

18.Brain Storm Optimization Algorithms with K-medians Clustering Algorithms

Author:Zhu, HY;Shi, YH

Source:2015 SEVENTH INTERNATIONAL CONFERENCE ON ADVANCED COMPUTATIONAL INTELLIGENCE (ICACI),2015,Vol.

Abstract:Brain storm optimization (BSO) algorithm is a novel swarm intelligence algorithm inspired by human beings' brainstorming process in problems solving. Generally, BSO algorithm has five main steps, which are initialization, evaluation, clustering, disruption and updating. In these five steps, the clustering step is critical to BSO algorithms. Original BSO algorithms use k-means methods as clustering algorithms, but k-means algorithm is affected by extreme values easily and the speed of algorithm is not high enough. In this paper, a variation of k-means clustering algorithm, called k-medians clustering algorithm, is investigated to replace k-means clustering algorithm. In addition, one modification is applied to both clustering algorithms, which is to replace the calculated cluster center with an individual closest to it. Experimental results show that the effectiveness of BSO does not change obviously, but the higher efficiency can be obtained.

19.Hierarchical Coordinated Control for DC Microgrid with Crowbar and Load Shedding Control

Author:Wen, HQ;Zheng, K;Du, Y

Source:2017 IEEE 3RD INTERNATIONAL FUTURE ENERGY ELECTRONICS CONFERENCE AND ECCE ASIA (IFEEC 2017-ECCE ASIA),2017,Vol.

Abstract:Due to the challenge of energy demand, renewable energy sources are introduced to our daily life in order to satisfy the energy demand and reduce the emission of greenhouse gas. But the increasing penetration of renewable energy sources causes obvious impact on the conventional power systems especially the control strategies. In order to solve the problem, microgrid is introduced, which usually consists of distributed generators(DG), storage system, loads and converters. There are two kinds of microgrid: DC microgrid and AC microgrid. Compare with AC microgrid, the conversion loss and transmission loss of DC microgrid is lower, furthermore, its control strategy is less complex. Due to the advantages of DC microgrid, it has drawn increasing attention in recent years especially the coordinated control of DC microgrid, which is very essential for the correct operation of DC microgrid. The control methods of DC microgrid can be divided into two categories: centralized control and distributed control. Since centralized control strongly relies on communication system, it gradually replaced by the distributed control, which can further classified into many subcategories such as droop control and hierarchical control. Compared with centralized control and droop control, hierarchical control method is easier and reduce the dependence of communication system, so the reliability of this control method can be ensured. In this paper, a hierarchical control strategy is provided to improve the operation performance of DC microgrid. The crowbar control and load shedding control are also included in this hierarchical control strategy. In this paper, the DC microgrid is set up in PSCAD and the hierarchical control is verified for different operating scenarios.

20.Reactive Power Loss Optimization Method for Bi-directional Isolated DC-DC Converters

Author:Wen, HQ

Source:2014 INTERNATIONAL POWER ELECTRONICS CONFERENCE (IPEC-HIROSHIMA 2014 - ECCE-ASIA),2014,Vol.

Abstract:In this paper, both the reactive power which delivered back to the source and from the load in the isolated bidirectional dual-active-bridge (DAB) dc-dc converter using the conventional phase-shift (CPS) control strategy are analyzed with mathematical expressions. An extended dual-phase-shift (EDPS) control is proposed and the reactive power losses are defined with regard to different operating cases. The experimental results reveal that the EDPS can effectively enhance the overall efficiency especially with the proposed method of reactive power losses minimization.
Total 185 results found
Copyright 2006-2020 © Xi'an Jiaotong-Liverpool University 苏ICP备07016150号-1 京公网安备 11010102002019号