Dear Editor,This letter is concerned with the role of recurrent neural networks(RNNs)on the controller design for a class of nonlinear systems.Inspired by the architectures of RNNs,the system states are stacked accord...Dear Editor,This letter is concerned with the role of recurrent neural networks(RNNs)on the controller design for a class of nonlinear systems.Inspired by the architectures of RNNs,the system states are stacked according to the dynamic along with time while the controller is represented as the neural network output.To build the bridge between RNNs and finite-time controller,a novel activation function is imposed on RNNs to drive the convergence of states at finite-time and propel the overall control process smoother.Rigorous stability proof is briefly provided for the convergence of the proposed finite-time controller.At last,a numerical simulation example is presented to illustrate the efficiency of the proposed strategy.Neural networks can be classified as static(feedforward)and dynamic(recurrent)nets[1].The former nets do not perform well in dealing with training data and using any information of the local data structure[2].In contrast to the feedforward neural networks,RNNs are constituted by high dimensional hidden states with dynamics.展开更多
Kernel learning forward backward stochastic differential equations(FBSDE)filter is an iterative and adaptive meshfree approach to solve the non-linear filtering problem.It builds from forward backward SDE for Fokker-P...Kernel learning forward backward stochastic differential equations(FBSDE)filter is an iterative and adaptive meshfree approach to solve the non-linear filtering problem.It builds from forward backward SDE for Fokker-Planker equation,which defines evolving density for the state variable,and employs kernel density estimation(KDE)to approximate density.This algo-rithm has shown more superior performance than mainstream particle filter method,in both convergence speed and efficiency of solving high dimension problems.However,this method has only been shown to converge empirically.In this paper,we present a rigorous analysis to demonstrate its local and global convergence,and provide theoretical support for its empirical results.展开更多
Second generation(2G)high-temperature superconductor(HTS)(RE)Ba2Cu3Ox(REBCO)shows a great potential in building high field magnets beyond 23.5 T.The electromagnetic modelling is vital for the design of HTS magnet,howe...Second generation(2G)high-temperature superconductor(HTS)(RE)Ba2Cu3Ox(REBCO)shows a great potential in building high field magnets beyond 23.5 T.The electromagnetic modelling is vital for the design of HTS magnet,however,this always suffers the challenge of huge computation for high field magnets with large number of turns.This study presents a novel electromagnetic modelling based on T-A formulation for REBCO magnets with thousands of turns.An equivalent turn method is proposed to reduce the number of turns in calculation,so that the computation cost can be reduced significantly,and meanwhile the key electromagnetic behaviour of HTS magnet can be simulated with enough accuracy.The ramping operation of a fully HTS magnet with 12,000 turns are analysed using both the original T-A model with actual turns and improved T-A model with equivalent turns.The two models show a good agreement on the key electromagnetic behaviours of the magnet:distribution of current density,magnetic fields,screen current induced field and magnetisation loss,so that this improved T-A model using equivalent turns is validated.The T-A modelling of REBCO magnet is a powerful tool for the electromagnetic analysis of industry-scale high field magnets.展开更多
Microwave Humidity Sounders （MHS） onboard NOAA- 15, - 16, - 17, - 18, - 19, and EUMETSAT MetOp-A/B satellites provide radiance measurements at a single polarization state at any of five observed frequen- cies. The M...Microwave Humidity Sounders （MHS） onboard NOAA- 15, - 16, - 17, - 18, - 19, and EUMETSAT MetOp-A/B satellites provide radiance measurements at a single polarization state at any of five observed frequen- cies. The Microwave Humidity Sounder （MWHS） onboard the FengYun-3 （FY-3） satellite has a unique instrument design that provides dual polarization measurements at 150 GHz. In this study, the MWHS polarization signal was investigated using observed and modeled data. It is shown that the quasi-polarization brightness temperatures at 150GHz display a scan angle dependent bias. Under calm ocean conditions, the polarization difference at 150 GHz becomes non-negligible when the scan angle varies from 10° to 45° and reaches a maximum when the scan angle is about 30°. Also, the polarization state is sensitive to surface parameters such as surface wind speed. Under clear-sky conditions, the differences between horizontal and vertical polarization states at 150GHz increase with decreasing surface wind speed. Therefore, the polarization signals from the cross-track scanning microwave measurements at window channels contain useful information about surface parameters. In addition, the availability of dual polarization measurements allows a one-to-one conversion from antenna brightness temperature to sensor brightness temperature ifa cross-polarization spill-over exists.展开更多
Silicone rubber composite is a priority electrical insulating material used in high-voltage outdoor insulation applications.Low electrical tracking/erosion and poor flame resistance performance of silicone rubber once...Silicone rubber composite is a priority electrical insulating material used in high-voltage outdoor insulation applications.Low electrical tracking/erosion and poor flame resistance performance of silicone rubber once ignited,substantially reduce its working life.This paper attempts to investigate tracking/erosion performance of room temperature vulcanized(RTV)silicone rubber along with flame retardant parameters using aluminum trihydrate(ATH),graphene nanosheets(GN)and milled glass fiber(GF)additives.The inclined plane test(IPT)was performed in line with criteria defined in IEC 60587 using step-up tracking voltage method while flame retardancy is evaluated according to ASTM E 1354.0 using a cone calorimeter.Results suggest 30% of ATH assists in improving physical tracking/erosion resistance of pristine silicone elastomer rubber by impeding development of leakage current and a great reduction in maximum average temperatures on the surface of RTV2.Further improvement in performance of RTV2 is achieved through introduction of 1%of GN and 5% of GF as seen in RTV4.Moreover,30% of ATH reduces heat release rate and smoke production rate,and this trend is improved with the introduction of GN/GF.RTV4 has pop up as the most promising silicone rubber composite with excellent electrical tracking,erosion,and flame resistance performance relative to its counterparts in this study.展开更多
文摘Dear Editor,This letter is concerned with the role of recurrent neural networks(RNNs)on the controller design for a class of nonlinear systems.Inspired by the architectures of RNNs,the system states are stacked according to the dynamic along with time while the controller is represented as the neural network output.To build the bridge between RNNs and finite-time controller,a novel activation function is imposed on RNNs to drive the convergence of states at finite-time and propel the overall control process smoother.Rigorous stability proof is briefly provided for the convergence of the proposed finite-time controller.At last,a numerical simulation example is presented to illustrate the efficiency of the proposed strategy.Neural networks can be classified as static(feedforward)and dynamic(recurrent)nets[1].The former nets do not perform well in dealing with training data and using any information of the local data structure[2].In contrast to the feedforward neural networks,RNNs are constituted by high dimensional hidden states with dynamics.
基金supported by the U.S.National Science Foundation through Project DMS-2142672by the U.S.Department of Energy,Office of Science,Office of Advanced Scientific Computing Research,Applied Mathematics Program under Grant DE-SC0022297.
文摘Kernel learning forward backward stochastic differential equations(FBSDE)filter is an iterative and adaptive meshfree approach to solve the non-linear filtering problem.It builds from forward backward SDE for Fokker-Planker equation,which defines evolving density for the state variable,and employs kernel density estimation(KDE)to approximate density.This algo-rithm has shown more superior performance than mainstream particle filter method,in both convergence speed and efficiency of solving high dimension problems.However,this method has only been shown to converge empirically.In this paper,we present a rigorous analysis to demonstrate its local and global convergence,and provide theoretical support for its empirical results.
基金the National Natural Science Foundation of China under grant nos.11802036 and 11872195the National High Magnetic Field Laboratory,which was supported by the U.S.National Science Foundation Cooperative Agreement no.DMR-1644779 and the State of Florida.
文摘Second generation(2G)high-temperature superconductor(HTS)(RE)Ba2Cu3Ox(REBCO)shows a great potential in building high field magnets beyond 23.5 T.The electromagnetic modelling is vital for the design of HTS magnet,however,this always suffers the challenge of huge computation for high field magnets with large number of turns.This study presents a novel electromagnetic modelling based on T-A formulation for REBCO magnets with thousands of turns.An equivalent turn method is proposed to reduce the number of turns in calculation,so that the computation cost can be reduced significantly,and meanwhile the key electromagnetic behaviour of HTS magnet can be simulated with enough accuracy.The ramping operation of a fully HTS magnet with 12,000 turns are analysed using both the original T-A model with actual turns and improved T-A model with equivalent turns.The two models show a good agreement on the key electromagnetic behaviours of the magnet:distribution of current density,magnetic fields,screen current induced field and magnetisation loss,so that this improved T-A model using equivalent turns is validated.The T-A modelling of REBCO magnet is a powerful tool for the electromagnetic analysis of industry-scale high field magnets.
文摘Microwave Humidity Sounders （MHS） onboard NOAA- 15, - 16, - 17, - 18, - 19, and EUMETSAT MetOp-A/B satellites provide radiance measurements at a single polarization state at any of five observed frequen- cies. The Microwave Humidity Sounder （MWHS） onboard the FengYun-3 （FY-3） satellite has a unique instrument design that provides dual polarization measurements at 150 GHz. In this study, the MWHS polarization signal was investigated using observed and modeled data. It is shown that the quasi-polarization brightness temperatures at 150GHz display a scan angle dependent bias. Under calm ocean conditions, the polarization difference at 150 GHz becomes non-negligible when the scan angle varies from 10° to 45° and reaches a maximum when the scan angle is about 30°. Also, the polarization state is sensitive to surface parameters such as surface wind speed. Under clear-sky conditions, the differences between horizontal and vertical polarization states at 150GHz increase with decreasing surface wind speed. Therefore, the polarization signals from the cross-track scanning microwave measurements at window channels contain useful information about surface parameters. In addition, the availability of dual polarization measurements allows a one-to-one conversion from antenna brightness temperature to sensor brightness temperature ifa cross-polarization spill-over exists.
基金financially through the Fulbright Postdoctoral-Vice Chancellor’s Postdoctoral Research Fellowship.
文摘Silicone rubber composite is a priority electrical insulating material used in high-voltage outdoor insulation applications.Low electrical tracking/erosion and poor flame resistance performance of silicone rubber once ignited,substantially reduce its working life.This paper attempts to investigate tracking/erosion performance of room temperature vulcanized(RTV)silicone rubber along with flame retardant parameters using aluminum trihydrate(ATH),graphene nanosheets(GN)and milled glass fiber(GF)additives.The inclined plane test(IPT)was performed in line with criteria defined in IEC 60587 using step-up tracking voltage method while flame retardancy is evaluated according to ASTM E 1354.0 using a cone calorimeter.Results suggest 30% of ATH assists in improving physical tracking/erosion resistance of pristine silicone elastomer rubber by impeding development of leakage current and a great reduction in maximum average temperatures on the surface of RTV2.Further improvement in performance of RTV2 is achieved through introduction of 1%of GN and 5% of GF as seen in RTV4.Moreover,30% of ATH reduces heat release rate and smoke production rate,and this trend is improved with the introduction of GN/GF.RTV4 has pop up as the most promising silicone rubber composite with excellent electrical tracking,erosion,and flame resistance performance relative to its counterparts in this study.
