A5100712049- Advanced Memory and Neural Computing
- Ferroelectric and Negative Capacitance Devices
- Power Systems and Renewable Energy
- High-Voltage Power Transmission Systems
- Smart Grid and Power Systems
- Semiconductor materials and devices
- Optical Network Technologies
- Photonic and Optical Devices
- Power Systems and Technologies
- Regional Economic and Spatial Analysis
- Advanced Photonic Communication Systems
- Power Systems Fault Detection
- Evaluation Methods in Various Fields
- Corporate Finance and Governance
- Neuroscience and Neural Engineering
- Semiconductor Lasers and Optical Devices
- Industrial Technology and Control Systems
- HVDC Systems and Fault Protection
- Real-time simulation and control systems
- Advanced MIMO Systems Optimization
- Extenics and Innovation Methods
- Collaboration in agile enterprises
- Optical Systems and Laser Technology
- Advanced Manufacturing and Logistics Optimization
- Advances in Cucurbitaceae Research
Chinese Academy of Sciences
2007-2025
Institute of Microelectronics
2021-2025
Marine Design & Research Institute of China
2025
University of Missouri–Kansas City
2015-2025
University of Chinese Academy of Sciences
2009-2024
Peng Cheng Laboratory
2024
Soochow University
2024
Tianjin Normal University
2010-2023
Sehan University
2023
Lanzhou Jiaotong University
2023
Resistive memory (RRAM) provides an ideal platform to develop embedded non-volatile computing-in-memory (nvCIM). However, it faces several critical challenges ranging from device non-idealities, large DC currents, and small signal margins. To address these issues, we propose voltage-division (VD) based computing approach its circuit implementation in two-transistor-two-resistor (2T2R) RRAM cell arrays, which can realize energy-efficient, sign-aware, robust deep neural network (DNN)...
The way media portray public health problems influences the public’s perception of and related solutions. Social allows users to engage with news collectively construct meaning. This paper examined in comparison user-generated content opioids understand role second-level agenda-setting health. We analyzed 162,760 tweets about opioid crisis, compared main topics their sentiments 2998 stories from New York Times online. Evidence this study suggests that agenda setting on social is different...
In this work, we present a four-transistor-two-resistor (4T2R) ternary content addressable memory (TCAM) bit cell based on the resistive (RRAM), comprising conventional two-transistor-two-resistor (2T2R) with two additional comparison transistors. It can effectively amplify match-line signal ratio (ML-ratio), lower leakage current of match (<inline-formula> <tex-math notation="LaTeX">${I}_{{\mathrm {MATCH}}}$ </tex-math></inline-formula>), and suppress read disturbance. The proposed concept...
Non-volatile computing-in-memory (nvCIM) can potentially meet the ever-increasing demands on improving energy efficiency (EF) for intelligent edge devices. However, it still suffers from limited input parallelism due to parasitic effects, signal margin degradation device non-idealities, and large hardware cost analog readout. In this work, we present a two-transistor-one-resistor (2T1R) resistive memory (RRAM) nvCIM macro featuring: 1) structure with decoupled computing data paths; 2)...
AI edge devices are not only required to perform inference tasks with low power and high real-time performance but also expected have the capability learn adapt dynamic unpredictable environments, without heavily relying on cloud-based training. The recent rise of computing-in-memory (CIM) has offered a competent solution by minimizing latency associated data movement. While many existing CIM designs [1–6] primarily focused improving inference, those learning abilities have, so far, been...
In "New Public Management" era, performance measurement has been widely used in managerial practices of public sectors. From the content and features measurement, this paper aims to explore inspirations on Chinese sector which based a review prior literatures including influencial factors, methods indicators evaluation. end, arguments are presented pointed out direction future researches field.
Micrometer-scale ITO convex matrixes are used to improve the light extraction efficiency of OLEDs. Through optical manipulation, this scheme enhances EQEs by up 34.5% for blue, 22.4% green, 28.6% red, and 31.3% NIR
In this work, we discussed the critical challenges, key enabling techniques, and future trends on developing RRAM-based brain-inspired computation, including computing-in-memory (CIM) neuromorphic computing (NC), from device, circuit to system. To suppress device non-idealities in synaptic array, proposed using optimized bit-cell design approach reduce errors power of analogue multiply-and-accumulate (MAC). lower neuron power, sparsity-aware analog-to-digital converter (ADC) for artificial...
This paper describes a MW-level island microgrid system which is constructed on 4226 meters high plateau in the western of China. The composed 7MWp PV modules, 5MWh lithium batteries and 20MWh lead-acid batteries, are integrated with power electronic interface without stable generation source. A master-slave control structure adopted system. Batteries Energy Storage Systems (BESSs) play as different function roles, cooperative method BESSs presented to support balance during operation....
With the rapid development of wind power, abandoned phenomenon is becoming increasingly serious. In order to study reasonable scale power in region, it necessary grid construction and operation cost different areas. this paper, optimal model established, which based on rate. The takes into account generation cost, penalty investment new line. generating unit includes starting stopping thermal turbine. Abandoned refers reduction with certain amount curtailment field punishment. line obtained...
In this paper, a low-complexity practical energy saving algorithm by switching off/on some eNBs in real dense urban scenario considering historical and real-time eNB load is proposed. First, are ranked according to their loads an ascending order the first list with decreasing smaller than low threshold pre-selected as target off cell. Then, effect of on neighbour evaluated. The switches-off while after switches another threshold. Since estimation additional due switch-off high complexity,...
Recent research on SDR (software defined radio) GNSS receiver is mostly based DSP/FPGA hybrid or PC platform. In this paper we will introduce a prototype design of in stand-alone DSP The with has more than 12 GPS channels, which implemented TI TMS320C6416 chip. chip fulfill acquisition, tracking and positioning function realtime. This method could be extended to other navigation constellation archive multi- onstellation receiver.
With the increasing of demand diversification and short product lifecycle, industries encounter challenges uncertainty. Japanese seru production system has received more attention with its efficiency flexibility. In this paper, a problem formulation under uncertain condition is researched. A multi-objective optimization model developed to minimize cost maximize service level system. The purpose paper formulate an energy which can response stochastic efficiently. Sample average approximation...
摘要: 以黄瓜(Cucumis sativus)子叶和子叶节为外植体, 利用农杆菌(Agrobacterium tumefaciens)介导法建立黄瓜遗传转化体系并进行优化, 将南方根结线虫(Meloidogyme incognita)寄生相关基因的RNA干扰载体导入黄瓜, 通过潮霉素(Hyg)浓度梯度筛选得到99株潮霉素抗性植株。经PCR和Southern-blot检测, 获得10株目的基因以单拷贝形式整合的转基因黄瓜, 子叶和子叶节的Southern-blot阳性率分别为13.9%和6.9%。该研究建立并优化了RNA干扰载体导入黄瓜的高效遗传转化体系, 成功获得了转基因黄瓜植株。
For HVDC converter station, a 12-pulse transformer is composed of six single-phase two-wing transformers generally, which are connected as Y0, y and d11 3-phase structure respectively. The two parallel accessed to AC bus. Combined with the building analyzing on electric equivalent circuit, theory researching mechanism characteristic zero-sequence inrush current energizing in detail, distribution principle laid out. With an failure event Zhengping station ±500kV Three-Gorges project,...
Resistive Memory (RRAM) based computing-in-memory (CIM) can play a key role in intermittently operated AI edge devices and sensors [1] –[3]. By minimizing the data movement during computation power switching, it effectively reduces hence improves real-time performance compared to conventional energy-efficient system. However, as shown Fig. 1, previous RRAM CIM macros, on 1T1R or 2T2R cell array, are still facing several critical challenges, including (1) large summation current induced...