Whole-angle gyroscopes have broad prospects for development with inherent advantages of excellent scale factor,wide bandwidth and measurement range,which are restrictions on rate gyroscopes.Previous studies on the who...Whole-angle gyroscopes have broad prospects for development with inherent advantages of excellent scale factor,wide bandwidth and measurement range,which are restrictions on rate gyroscopes.Previous studies on the whole-angle mode are based mostly on the linear model of Lynch,and the essential nonlinearity of capacitive displacement detection is always neglected,which has significant negative effects on the performance.In this paper,a novel realtime calibration method of capacitive displacement detection is proposed to eliminate these nonlinear effects.This novel method innovatively takes advantage of the relationship between the first and third harmonic components of detective signals for calibration.Based on this method,the real-time calibration of capacitive displacement detection is achieved and solves the problems of traditional methods,which are usually related to the vibration amplitude,environmental variations and other factors.Furthermore,this novel calibration method is embedded into a whole-angle control system to restore the linear capacitive response in real time and then combined with a microshell resonator for the first time to exploit the enormous potential of an ultrahigh Q factor and symmetric structure.The effectiveness is proven because the angle drift is reduced significantly to improve the scale-factor nonlinearity by 14 times to 0.79 ppm with 0.0673/h bias instability and a 0.0017s rate threshold,which is the best reported performance of the MEMS whole-angle gyroscope thus far.More importantly,this novel calibration method can be applied for all kinds of resonators with the requirement of a linear capacitive response even under a large resonant amplitude.展开更多
Fraud cases have been a risk in society and people’s property security has been greatly threatened.In recent studies,many promising algorithms have been developed for social media offensive text recognition as well a...Fraud cases have been a risk in society and people’s property security has been greatly threatened.In recent studies,many promising algorithms have been developed for social media offensive text recognition as well as sentiment analysis.These algorithms are also suitable for fraudulent phone text recognition.Compared to these tasks,the semantics of fraudulent words are more complex and more difficult to distinguish.Recurrent Neural Networks(RNN),the variants ofRNN,ConvolutionalNeuralNetworks(CNN),and hybrid neural networks to extract text features are used by most text classification research.However,a single network or a simple network combination cannot obtain rich characteristic knowledge of fraudulent phone texts relatively.Therefore,a new model is proposed in this paper.In the fraudulent phone text,the knowledge that can be learned by the model includes the sequence structure of sentences,the correlation between words,the correlation of contextual semantics,the feature of keywords in sentences,etc.The new model combines a bidirectional Long-Short Term Memory Neural Network(BiLSTM)or a bidirectional Gate Recurrent United(BiGRU)and a Multi-Head attention mechanism module with convolution.A normalization layer is added after the output of the final hidden layer.BiLSTM or BiGRU is used to build the encoding and decoding layer.Multi-head attention mechanism module with convolution(MHAC)enhances the ability of the model to learn global interaction information and multi-granularity local interaction information in fraudulent sentences.A fraudulent phone text dataset is produced by us in this paper.The THUCNews data sets and fraudulent phone text data sets are used in experiments.Experiment results show that compared with the baseline model,the proposed model(LMHACL)has the best experiment results in terms of Accuracy,Precision,Recall,and F1 score on the two data sets.And the performance indexes on fraudulent phone text data sets are all above 0.94.展开更多
Bismuth(Bi)-doped photonic materials, which exhibit broadband near-infrared(NIR) luminescence(1000–1600 nm), are evolving into interesting gain media. However, the traditional methods have shown their limitations in ...Bismuth(Bi)-doped photonic materials, which exhibit broadband near-infrared(NIR) luminescence(1000–1600 nm), are evolving into interesting gain media. However, the traditional methods have shown their limitations in enhancing Bi NIR emission, especially in the microregion. Consequently, the typical NIR emission has seldom been achieved in Bi-doped waveguides, which highly restricts the application of Bi-activated materials.Here, superbroadband Bi NIR emission is induced in situ instantly in the grating region by a femtosecond(fs)laser inside borosilicate glasses. A series of structural and spectroscopic characterizations are summoned to probe the generation mechanism. And we show how this novel NIR emission in the grating region can be enhanced significantly and erased reversibly. Furthermore, we successfully demonstrate Bi-activated optical waveguides.These results present new insights into Bi-doped materials and push the development of broadband waveguide amplification.展开更多
基金the National Natural Science Foundation of China under Grant 51935013,52075540,51905538the National Key R&D Program of China under Grant 2018YFB2002304The Hunan Science Foundation for Distinguished Young Scholars 2020JJ2033 is also acknowledged gratefully。
文摘Whole-angle gyroscopes have broad prospects for development with inherent advantages of excellent scale factor,wide bandwidth and measurement range,which are restrictions on rate gyroscopes.Previous studies on the whole-angle mode are based mostly on the linear model of Lynch,and the essential nonlinearity of capacitive displacement detection is always neglected,which has significant negative effects on the performance.In this paper,a novel realtime calibration method of capacitive displacement detection is proposed to eliminate these nonlinear effects.This novel method innovatively takes advantage of the relationship between the first and third harmonic components of detective signals for calibration.Based on this method,the real-time calibration of capacitive displacement detection is achieved and solves the problems of traditional methods,which are usually related to the vibration amplitude,environmental variations and other factors.Furthermore,this novel calibration method is embedded into a whole-angle control system to restore the linear capacitive response in real time and then combined with a microshell resonator for the first time to exploit the enormous potential of an ultrahigh Q factor and symmetric structure.The effectiveness is proven because the angle drift is reduced significantly to improve the scale-factor nonlinearity by 14 times to 0.79 ppm with 0.0673/h bias instability and a 0.0017s rate threshold,which is the best reported performance of the MEMS whole-angle gyroscope thus far.More importantly,this novel calibration method can be applied for all kinds of resonators with the requirement of a linear capacitive response even under a large resonant amplitude.
基金This researchwas funded by the Major Science and Technology Innovation Project of Shandong Province in China(2019JZZY010120).
文摘Fraud cases have been a risk in society and people’s property security has been greatly threatened.In recent studies,many promising algorithms have been developed for social media offensive text recognition as well as sentiment analysis.These algorithms are also suitable for fraudulent phone text recognition.Compared to these tasks,the semantics of fraudulent words are more complex and more difficult to distinguish.Recurrent Neural Networks(RNN),the variants ofRNN,ConvolutionalNeuralNetworks(CNN),and hybrid neural networks to extract text features are used by most text classification research.However,a single network or a simple network combination cannot obtain rich characteristic knowledge of fraudulent phone texts relatively.Therefore,a new model is proposed in this paper.In the fraudulent phone text,the knowledge that can be learned by the model includes the sequence structure of sentences,the correlation between words,the correlation of contextual semantics,the feature of keywords in sentences,etc.The new model combines a bidirectional Long-Short Term Memory Neural Network(BiLSTM)or a bidirectional Gate Recurrent United(BiGRU)and a Multi-Head attention mechanism module with convolution.A normalization layer is added after the output of the final hidden layer.BiLSTM or BiGRU is used to build the encoding and decoding layer.Multi-head attention mechanism module with convolution(MHAC)enhances the ability of the model to learn global interaction information and multi-granularity local interaction information in fraudulent sentences.A fraudulent phone text dataset is produced by us in this paper.The THUCNews data sets and fraudulent phone text data sets are used in experiments.Experiment results show that compared with the baseline model,the proposed model(LMHACL)has the best experiment results in terms of Accuracy,Precision,Recall,and F1 score on the two data sets.And the performance indexes on fraudulent phone text data sets are all above 0.94.
基金Natural Science Foundation of Guangdong Province(2018B030308009)National Natural Science Foundation of China(NSFC)(51672085)+3 种基金Program for Innovative Research Team in University of Ministry of Education of China(IRT_17R38)Ministry of Education of the People's Republic of China(MOE)Local Innovative Research Team Project of "Pearl River Talent Plan"(2017BT01X137)Fundamental Research Funds for the Central Universities
文摘Bismuth(Bi)-doped photonic materials, which exhibit broadband near-infrared(NIR) luminescence(1000–1600 nm), are evolving into interesting gain media. However, the traditional methods have shown their limitations in enhancing Bi NIR emission, especially in the microregion. Consequently, the typical NIR emission has seldom been achieved in Bi-doped waveguides, which highly restricts the application of Bi-activated materials.Here, superbroadband Bi NIR emission is induced in situ instantly in the grating region by a femtosecond(fs)laser inside borosilicate glasses. A series of structural and spectroscopic characterizations are summoned to probe the generation mechanism. And we show how this novel NIR emission in the grating region can be enhanced significantly and erased reversibly. Furthermore, we successfully demonstrate Bi-activated optical waveguides.These results present new insights into Bi-doped materials and push the development of broadband waveguide amplification.
