The reservoir computing(RC)system,known for its ability to seamlessly integrate memory and computing functions,is considered as a promising solution to meet the high demands for time and energy-efficient computing in ...The reservoir computing(RC)system,known for its ability to seamlessly integrate memory and computing functions,is considered as a promising solution to meet the high demands for time and energy-efficient computing in the current big data landscape,compared with traditional silicon-based computing systems that have a noticeable disadvantage of separate storage and computation.This review focuses on in-materio RC based on nanowire networks(NWs)from the perspective of materials,extending to reservoir devices and applications.The common methods used in preparing nanowires-based reservoirs,including the synthesis of nanowires and the construction of networks,are firstly systematically summarized.The physical principles of memristive and memcapacitive junctions are then explained.Afterwards,the dynamic characteristics of nanowires-based reservoirs and their computing capability,as well as the neuromorphic applications of NWs-based RC systems in recognition,classification,and forecasting tasks,are explicated in detail.Lastly,the current challenges and future opportunities facing NWs-based RC are highlighted,aiming to provide guidance for further research.展开更多
Text extraction from images using the traditional techniques of image collecting,and pattern recognition using machine learning consume time due to the amount of extracted features from the images.Deep Neural Networks...Text extraction from images using the traditional techniques of image collecting,and pattern recognition using machine learning consume time due to the amount of extracted features from the images.Deep Neural Networks introduce effective solutions to extract text features from images using a few techniques and the ability to train large datasets of images with significant results.This study proposes using Dual Maxpooling and concatenating convolution Neural Networks(CNN)layers with the activation functions Relu and the Optimized Leaky Relu(OLRelu).The proposed method works by dividing the word image into slices that contain characters.Then pass them to deep learning layers to extract feature maps and reform the predicted words.Bidirectional Short Memory(BiLSTM)layers extractmore compelling features and link the time sequence fromforward and backward directions during the training phase.The Connectionist Temporal Classification(CTC)function calcifies the training and validation loss rates.In addition to decoding the extracted feature to reform characters again and linking them according to their time sequence.The proposed model performance is evaluated using training and validation loss errors on the Mjsynth and Integrated Argument Mining Tasks(IAM)datasets.The result of IAM was 2.09%for the average loss errors with the proposed dualMaxpooling and OLRelu.In the Mjsynth dataset,the best validation loss rate shrunk to 2.2%by applying concatenating CNN layers,and Relu.展开更多
In the present study,the thermal,mechanical,and biological properties of xAg/Ti-30Ta(x=0,0.41,0.82 and 2.48 at%)shape memory alloys(SMAs)were investigated.The study was conducted using optical and scanning electron mi...In the present study,the thermal,mechanical,and biological properties of xAg/Ti-30Ta(x=0,0.41,0.82 and 2.48 at%)shape memory alloys(SMAs)were investigated.The study was conducted using optical and scanning electron microscopy(SEM),X-ray diffractometry(XRD),compression test,and shape memory testing.The xAg/Ti-Ta was made using a powder metallurgy technique and microwave-sintering process.The results revealed that the addition of Ag has a significant effect on the pore size and shape,whereas the smallest pore size of 11μm was found with the addition of 0.41 at%along with a relative density of 72%.The fracture stress and strain increased with the addition of Ag,reaching the minimum values around 0.41 at%Ag.Therefore,this composition showed the maximum stress and strain at fracture region.Moreover,0.82 Ag/Ti-Ta shows more excellent corrosion resistance and biocompatibility than other percentages,obtaining almost the same behaviour of the pure Ti and Ti-6Al-4V alloys,which can be recommended for their promising and potential response for biomaterial applications.展开更多
选择了位于华东地区的某医疗废物焚烧处置设施开展启炉过程与正常工况下烟气和飞灰的二英排放特性对比研究。数据显示,启炉后期烟气中二英的浓度达到1.68 ng I-TEQ·m-3,在焚烧炉温度稳定以后12 h,达到2.77 ng I-TEQ·m-3,飞灰...选择了位于华东地区的某医疗废物焚烧处置设施开展启炉过程与正常工况下烟气和飞灰的二英排放特性对比研究。数据显示,启炉后期烟气中二英的浓度达到1.68 ng I-TEQ·m-3,在焚烧炉温度稳定以后12 h,达到2.77 ng I-TEQ·m-3,飞灰中二英毒性当量水平也达到4.5 ng I-TEQ·g-1。启炉过程中烟气中气相二英所占比例逐渐增加,从平均占到50%增加到超过90%。启炉过程中烟气二英排放速率高于其正常达标时的排放水平,最高值为58.1μg I-TEQ·h-1,超过正常排放的40倍。一个启炉周期二英的排放总量达到0.785 mg I-TEQ,达标正常工况下二英的年排放总量为8.4 mg I-TEQ,按照平均每年3次启炉来计算,启炉过程二英的排放量占到全年正常排放的28%。展开更多
： Metal-oxide-nitride-oxide-silicon （MONOS） capacitors with thermally grown SiO2 as the tunnel layer are fabricated, and the effects of different ambient nitridation （NH3, NO and N20） on the characteristics of th...： Metal-oxide-nitride-oxide-silicon （MONOS） capacitors with thermally grown SiO2 as the tunnel layer are fabricated, and the effects of different ambient nitridation （NH3, NO and N20） on the characteristics of the memory capacitors are investigated. The experimental results indicate that the device with tunnel oxide annealed in NO ambient exhibits excellent memory characteristics, i.e. a large memory window, high program/erase speed, and good endurance and retention performance （the charge loss rate is 14.5% after l0 years）. The mechanism involved is that much more nitrogen is incorporated into the tunnel oxide during NO annealing, resulting in a lower tunneling barrier height and smaller interface state density. Thus, there is a higher tunneling rate under a high electric field and a lower probability of trap-assisted tunneling during retention, as compared to N20 annealing. Furthermore, compared with the NH3-annealed device, no weak Si-H bonds and electron traps related to the hydrogen are introduced for the NO-annealed devices, giving a high-quality and high-reliability SiON tunneling layer and SiON/Si interface due to the suitable nitridation and oxidation roles of NO. Key words： MONOS memory; memory characteristics; annealing; nitridation展开更多
基金financially supported by the National Key R&D Program of China(Grant No.2020AAA0109005)the Strategy Priority Research Program of Chinese Academy of Sciences(Grant No.XDA0330100)+1 种基金the Beijing Municipal Science&Technology Commission Program of China(Grant No.Z201100004320004)the China Association for Science and Technology(Grant No.2019Q1NRC001).
文摘The reservoir computing(RC)system,known for its ability to seamlessly integrate memory and computing functions,is considered as a promising solution to meet the high demands for time and energy-efficient computing in the current big data landscape,compared with traditional silicon-based computing systems that have a noticeable disadvantage of separate storage and computation.This review focuses on in-materio RC based on nanowire networks(NWs)from the perspective of materials,extending to reservoir devices and applications.The common methods used in preparing nanowires-based reservoirs,including the synthesis of nanowires and the construction of networks,are firstly systematically summarized.The physical principles of memristive and memcapacitive junctions are then explained.Afterwards,the dynamic characteristics of nanowires-based reservoirs and their computing capability,as well as the neuromorphic applications of NWs-based RC systems in recognition,classification,and forecasting tasks,are explicated in detail.Lastly,the current challenges and future opportunities facing NWs-based RC are highlighted,aiming to provide guidance for further research.
基金supported this project under the Fundamental Research Grant Scheme(FRGS)FRGS/1/2019/ICT02/UKM/02/9 entitled“Convolution Neural Network Enhancement Based on Adaptive Convexity and Regularization Functions for Fake Video Analytics”.This grant was received by Prof.Assis.Dr.S.N.H.Sheikh Abdullah,https://www.ukm.my/spifper/research_news/instrumentfunds.
文摘Text extraction from images using the traditional techniques of image collecting,and pattern recognition using machine learning consume time due to the amount of extracted features from the images.Deep Neural Networks introduce effective solutions to extract text features from images using a few techniques and the ability to train large datasets of images with significant results.This study proposes using Dual Maxpooling and concatenating convolution Neural Networks(CNN)layers with the activation functions Relu and the Optimized Leaky Relu(OLRelu).The proposed method works by dividing the word image into slices that contain characters.Then pass them to deep learning layers to extract feature maps and reform the predicted words.Bidirectional Short Memory(BiLSTM)layers extractmore compelling features and link the time sequence fromforward and backward directions during the training phase.The Connectionist Temporal Classification(CTC)function calcifies the training and validation loss rates.In addition to decoding the extracted feature to reform characters again and linking them according to their time sequence.The proposed model performance is evaluated using training and validation loss errors on the Mjsynth and Integrated Argument Mining Tasks(IAM)datasets.The result of IAM was 2.09%for the average loss errors with the proposed dualMaxpooling and OLRelu.In the Mjsynth dataset,the best validation loss rate shrunk to 2.2%by applying concatenating CNN layers,and Relu.
基金Project(Q.J130000.2524.12H60)supported by the Ministry of Higher Education of Malaysia and Universiti Teknologi Malaysia。
文摘In the present study,the thermal,mechanical,and biological properties of xAg/Ti-30Ta(x=0,0.41,0.82 and 2.48 at%)shape memory alloys(SMAs)were investigated.The study was conducted using optical and scanning electron microscopy(SEM),X-ray diffractometry(XRD),compression test,and shape memory testing.The xAg/Ti-Ta was made using a powder metallurgy technique and microwave-sintering process.The results revealed that the addition of Ag has a significant effect on the pore size and shape,whereas the smallest pore size of 11μm was found with the addition of 0.41 at%along with a relative density of 72%.The fracture stress and strain increased with the addition of Ag,reaching the minimum values around 0.41 at%Ag.Therefore,this composition showed the maximum stress and strain at fracture region.Moreover,0.82 Ag/Ti-Ta shows more excellent corrosion resistance and biocompatibility than other percentages,obtaining almost the same behaviour of the pure Ti and Ti-6Al-4V alloys,which can be recommended for their promising and potential response for biomaterial applications.
文摘选择了位于华东地区的某医疗废物焚烧处置设施开展启炉过程与正常工况下烟气和飞灰的二英排放特性对比研究。数据显示,启炉后期烟气中二英的浓度达到1.68 ng I-TEQ·m-3,在焚烧炉温度稳定以后12 h,达到2.77 ng I-TEQ·m-3,飞灰中二英毒性当量水平也达到4.5 ng I-TEQ·g-1。启炉过程中烟气中气相二英所占比例逐渐增加,从平均占到50%增加到超过90%。启炉过程中烟气二英排放速率高于其正常达标时的排放水平,最高值为58.1μg I-TEQ·h-1,超过正常排放的40倍。一个启炉周期二英的排放总量达到0.785 mg I-TEQ,达标正常工况下二英的年排放总量为8.4 mg I-TEQ,按照平均每年3次启炉来计算,启炉过程二英的排放量占到全年正常排放的28%。
基金supported by the National Natural Science Foundation of China(No.60976091)
文摘： Metal-oxide-nitride-oxide-silicon （MONOS） capacitors with thermally grown SiO2 as the tunnel layer are fabricated, and the effects of different ambient nitridation （NH3, NO and N20） on the characteristics of the memory capacitors are investigated. The experimental results indicate that the device with tunnel oxide annealed in NO ambient exhibits excellent memory characteristics, i.e. a large memory window, high program/erase speed, and good endurance and retention performance （the charge loss rate is 14.5% after l0 years）. The mechanism involved is that much more nitrogen is incorporated into the tunnel oxide during NO annealing, resulting in a lower tunneling barrier height and smaller interface state density. Thus, there is a higher tunneling rate under a high electric field and a lower probability of trap-assisted tunneling during retention, as compared to N20 annealing. Furthermore, compared with the NH3-annealed device, no weak Si-H bonds and electron traps related to the hydrogen are introduced for the NO-annealed devices, giving a high-quality and high-reliability SiON tunneling layer and SiON/Si interface due to the suitable nitridation and oxidation roles of NO. Key words： MONOS memory; memory characteristics; annealing; nitridation
