We consider solving the forward and inverse partial differential equations(PDEs)which have sharp solutions with physics-informed neural networks(PINNs)in this work.In particular,to better capture the sharpness of the ...We consider solving the forward and inverse partial differential equations(PDEs)which have sharp solutions with physics-informed neural networks(PINNs)in this work.In particular,to better capture the sharpness of the solution,we propose the adaptive sampling methods(ASMs)based on the residual and the gradient of the solution.We first present a residual only-based ASM denoted by ASMⅠ.In this approach,we first train the neural network using a small number of residual points and divide the computational domain into a certain number of sub-domains,then we add new residual points in the sub-domain which has the largest mean absolute value of the residual,and those points which have the largest absolute values of the residual in this sub-domain as new residual points.We further develop a second type of ASM(denoted by ASMⅡ)based on both the residual and the gradient of the solution due to the fact that only the residual may not be able to efficiently capture the sharpness of the solution.The procedure of ASMⅡis almost the same as that of ASMⅠ,and we add new residual points which have not only large residuals but also large gradients.To demonstrate the effectiveness of the present methods,we use both ASMⅠand ASMⅡto solve a number of PDEs,including the Burger equation,the compressible Euler equation,the Poisson equation over an Lshape domain as well as the high-dimensional Poisson equation.It has been shown from the numerical results that the sharp solutions can be well approximated by using either ASMⅠor ASMⅡ,and both methods deliver much more accurate solutions than the original PINNs with the same number of residual points.Moreover,the ASMⅡalgorithm has better performance in terms of accuracy,efficiency,and stability compared with the ASMⅠalgorithm.This means that the gradient of the solution improves the stability and efficiency of the adaptive sampling procedure as well as the accuracy of the solution.Furthermore,we also employ the similar adaptive sampling technique for the data points of boundar展开更多
This paper represents the performance analysis of the different shapes of antenna balun and feeding techniques for step constant tapered slot antenna. This work also addresses the benefits of antenna balun (circular a...This paper represents the performance analysis of the different shapes of antenna balun and feeding techniques for step constant tapered slot antenna. This work also addresses the benefits of antenna balun (circular and rectangular) along with two types of feeding techniques (Microstrip line L-shape and Microstrip line I-shape). The performance of the antenna for each technique is thoroughly investigated using Computer Simulation Technology (CST) Microwave Studio software simulation under the resonant frequency of 5.9 GHz. Results demonstrate that the proposed model is an effective tool for improving antenna performance. Moreover, an extensive comparison has been carried out between the two different shapes, with and without antenna balun and between two feeding techniques focusing on return losses, gain, directivity, and voltage standing wave ratio (VSWR).展开更多
The basic scheme of the orientation detection system using L-shape reticle is introduced. The dimension of the patterns on the reticle of the system in practical applications is designed and an analysis of the princip...The basic scheme of the orientation detection system using L-shape reticle is introduced. The dimension of the patterns on the reticle of the system in practical applications is designed and an analysis of the principle of abstracting the orientation information of the target and the effects and formation method of self-adapting tracking gate is presented. The research result shows that the orientation detection system using L-shape reticle has a good effect on space-filtering, the signals that the orientation detection system sends out are easy to be processed by computer, its self-adapting tracking gate has a strong anti-interference ability, and the whole system＇s searching and tracking performances are quite high.展开更多
A new method is presented to estimate two-dimensional （2-D） Direction-of-Arrival （DOA） angles of narrowband real-valued signals impinging on a L-shape Arrays（LA）. The basic idea of the proposed method is to incr...A new method is presented to estimate two-dimensional （2-D） Direction-of-Arrival （DOA） angles of narrowband real-valued signals impinging on a L-shape Arrays（LA）. The basic idea of the proposed method is to increase both the effective aperture size and the number of sensors by employing the conjugate invariance property of real-valued signals. Thus, the proposed method can provide a more precise DOA and detect more signals than the Cross-Correlation Matrix Method （CCMM）. Numerical simulation results are presented to support the theory.展开更多
Effects of layer quantities and stacking sequences on L-shape composite manufacturing qualities in using OOA(out-of-autoclave)prepregs were studied.The mechanisms of air evacuated in 5 kinds of lay-ups were revealed b...Effects of layer quantities and stacking sequences on L-shape composite manufacturing qualities in using OOA(out-of-autoclave)prepregs were studied.The mechanisms of air evacuated in 5 kinds of lay-ups were revealed by image analysis of cut surfaces and thickness measurements.Results show that air in OOA prepregs is evacuated in two ways.Most of the air is forced out of layers directly by vacuum before air accesses in prepregs closed.Very little entrapped air moves perpendicularly to outer layers under hydrostatic resin pressure.When a laminate contains less than 16 layers,voids can hardly be found in layers.When a laminate contains more than 16 layers,voids cannot be expelled completely during the window of vertical movement.As for stacking sequences,the synergetic effect of slip function and nest function determines the thickness and voids content of laminates.Results show that the average of single layer thickness of unidirectional layers is the lowest,and the average of single layer thickness of quasi-isotropic layers is the highest.The voids content of quasi isotropic is the highest,which is consistent with the theoretical analysis.展开更多
基金Project supported by the National Key R&D Program of China(No.2022YFA1004504)the National Natural Science Foundation of China(Nos.12171404 and 12201229)the Fundamental Research Funds for Central Universities of China(No.20720210037)。
文摘We consider solving the forward and inverse partial differential equations(PDEs)which have sharp solutions with physics-informed neural networks(PINNs)in this work.In particular,to better capture the sharpness of the solution,we propose the adaptive sampling methods(ASMs)based on the residual and the gradient of the solution.We first present a residual only-based ASM denoted by ASMⅠ.In this approach,we first train the neural network using a small number of residual points and divide the computational domain into a certain number of sub-domains,then we add new residual points in the sub-domain which has the largest mean absolute value of the residual,and those points which have the largest absolute values of the residual in this sub-domain as new residual points.We further develop a second type of ASM(denoted by ASMⅡ)based on both the residual and the gradient of the solution due to the fact that only the residual may not be able to efficiently capture the sharpness of the solution.The procedure of ASMⅡis almost the same as that of ASMⅠ,and we add new residual points which have not only large residuals but also large gradients.To demonstrate the effectiveness of the present methods,we use both ASMⅠand ASMⅡto solve a number of PDEs,including the Burger equation,the compressible Euler equation,the Poisson equation over an Lshape domain as well as the high-dimensional Poisson equation.It has been shown from the numerical results that the sharp solutions can be well approximated by using either ASMⅠor ASMⅡ,and both methods deliver much more accurate solutions than the original PINNs with the same number of residual points.Moreover,the ASMⅡalgorithm has better performance in terms of accuracy,efficiency,and stability compared with the ASMⅠalgorithm.This means that the gradient of the solution improves the stability and efficiency of the adaptive sampling procedure as well as the accuracy of the solution.Furthermore,we also employ the similar adaptive sampling technique for the data points of boundar
文摘This paper represents the performance analysis of the different shapes of antenna balun and feeding techniques for step constant tapered slot antenna. This work also addresses the benefits of antenna balun (circular and rectangular) along with two types of feeding techniques (Microstrip line L-shape and Microstrip line I-shape). The performance of the antenna for each technique is thoroughly investigated using Computer Simulation Technology (CST) Microwave Studio software simulation under the resonant frequency of 5.9 GHz. Results demonstrate that the proposed model is an effective tool for improving antenna performance. Moreover, an extensive comparison has been carried out between the two different shapes, with and without antenna balun and between two feeding techniques focusing on return losses, gain, directivity, and voltage standing wave ratio (VSWR).
文摘The basic scheme of the orientation detection system using L-shape reticle is introduced. The dimension of the patterns on the reticle of the system in practical applications is designed and an analysis of the principle of abstracting the orientation information of the target and the effects and formation method of self-adapting tracking gate is presented. The research result shows that the orientation detection system using L-shape reticle has a good effect on space-filtering, the signals that the orientation detection system sends out are easy to be processed by computer, its self-adapting tracking gate has a strong anti-interference ability, and the whole system＇s searching and tracking performances are quite high.
基金Supported by Program for New Century Excellent Talents in University
文摘A new method is presented to estimate two-dimensional （2-D） Direction-of-Arrival （DOA） angles of narrowband real-valued signals impinging on a L-shape Arrays（LA）. The basic idea of the proposed method is to increase both the effective aperture size and the number of sensors by employing the conjugate invariance property of real-valued signals. Thus, the proposed method can provide a more precise DOA and detect more signals than the Cross-Correlation Matrix Method （CCMM）. Numerical simulation results are presented to support the theory.
文摘Effects of layer quantities and stacking sequences on L-shape composite manufacturing qualities in using OOA(out-of-autoclave)prepregs were studied.The mechanisms of air evacuated in 5 kinds of lay-ups were revealed by image analysis of cut surfaces and thickness measurements.Results show that air in OOA prepregs is evacuated in two ways.Most of the air is forced out of layers directly by vacuum before air accesses in prepregs closed.Very little entrapped air moves perpendicularly to outer layers under hydrostatic resin pressure.When a laminate contains less than 16 layers,voids can hardly be found in layers.When a laminate contains more than 16 layers,voids cannot be expelled completely during the window of vertical movement.As for stacking sequences,the synergetic effect of slip function and nest function determines the thickness and voids content of laminates.Results show that the average of single layer thickness of unidirectional layers is the lowest,and the average of single layer thickness of quasi-isotropic layers is the highest.The voids content of quasi isotropic is the highest,which is consistent with the theoretical analysis.
