Transactions of Nanjing University of Aeronautics & Astronautics
Issue 3,2010 Table of Contents
2010(3):199-204.
Abstract:
An adaptive method for the solution of compressible flows is described. The idea results from the desire for an efficient grid system, and an accurate and robust solution method are used to resolve flow features of the interest. The adaptation flow solution is proposed, including the detection of flow features based on the matrix error; the mesh adaptation using the mesh movement, the mesh refinement, the mesh coarsening, and their combination. The feature detection based on the matrix error can maintain the high olution property for shock waves of the onedimensional approximate Riemann solver and the higher order reconstruction scheme. The high grid efficiency is obtained with the anisotropically directional grid corresponding to feature directions, and the error of the flow field is averaged. The procedure and its application to flow solutions of shock waves are described. Results validate that the method is reliable.
An adaptive method for the solution of compressible flows is described. The idea results from the desire for an efficient grid system, and an accurate and robust solution method are used to resolve flow features of the interest. The adaptation flow solution is proposed, including the detection of flow features based on the matrix error; the mesh adaptation using the mesh movement, the mesh refinement, the mesh coarsening, and their combination. The feature detection based on the matrix error can maintain the high olution property for shock waves of the onedimensional approximate Riemann solver and the higher order reconstruction scheme. The high grid efficiency is obtained with the anisotropically directional grid corresponding to feature directions, and the error of the flow field is averaged. The procedure and its application to flow solutions of shock waves are described. Results validate that the method is reliable.
2010(3):205-212.
Abstract:
To calculate the aerodynamics of flapping wing micro air vehicle(MAV) with the high efficiency and the engineering oriented accuracy, an improved unsteady vortex lattice method (UVLM) for MAV is proposed. The method considers the influence of instantaneous wing deforming in flapping, as well as the induced drag, additionally models the stretching and the dissipation of vortex rings, and can present the aerodynamics status on the wing surface. An implementation of the method is developed. Moreover, the results and the efficiency of the proposed method are verified by CFD methods. Considering the less time cost of UVLM,for application of UVLM in the MAV optimization, the influence of wake vortex ignoring time saving and precision is studied. Results show that saving in CPU time with wake vortex ignoring the appropriate distance is considerable while the precision is not significantly reduced. It indicates the potential value of UVLM in the optimization of MAV design.
To calculate the aerodynamics of flapping wing micro air vehicle(MAV) with the high efficiency and the engineering oriented accuracy, an improved unsteady vortex lattice method (UVLM) for MAV is proposed. The method considers the influence of instantaneous wing deforming in flapping, as well as the induced drag, additionally models the stretching and the dissipation of vortex rings, and can present the aerodynamics status on the wing surface. An implementation of the method is developed. Moreover, the results and the efficiency of the proposed method are verified by CFD methods. Considering the less time cost of UVLM,for application of UVLM in the MAV optimization, the influence of wake vortex ignoring time saving and precision is studied. Results show that saving in CPU time with wake vortex ignoring the appropriate distance is considerable while the precision is not significantly reduced. It indicates the potential value of UVLM in the optimization of MAV design.
2010(3):213-218.
Abstract:
A new reliability based multidisciplinary design optimization (RBMDO) framework is proposed by combining the single loop based reliability analysis (SLBRA) method with multidisciplinary feasible (MDF) method. The Kriging approximate model with updating is introduced to reduce the computational cost of MDF caused by the complex structure. The computational efficiency is remarkably improved as the lack of iterative process during reliability analysis. Special attention is paid to a turbine blade design optimization by adopting the proposed method. Results show that the method is much more efficient than the commonly used double loop based RBMDO method. It is feasible and efficient to apply the method to the engineering design.
A new reliability based multidisciplinary design optimization (RBMDO) framework is proposed by combining the single loop based reliability analysis (SLBRA) method with multidisciplinary feasible (MDF) method. The Kriging approximate model with updating is introduced to reduce the computational cost of MDF caused by the complex structure. The computational efficiency is remarkably improved as the lack of iterative process during reliability analysis. Special attention is paid to a turbine blade design optimization by adopting the proposed method. Results show that the method is much more efficient than the commonly used double loop based RBMDO method. It is feasible and efficient to apply the method to the engineering design.
2010(3):219-225.
Abstract:
Aimed at the complex flight control system of a fighter, a kind of robust adapti ve control methods using multiple models is presented to make the control system track the given signal under different working conditions and to reconfigure th e control law for some structural failures. Firstly, the multiplemodel control structure is formed by several linear models and one fuzzy model. In the fuzzy logic way, weights of the multiplemodel adaptive controller are obtained. Then , a dynamic structure adaptive neural network is introduced to stabilize the who le system and eliminate the influence caused by the frequent switching. Simulati on results show that the control method is effective by demonstrating the normal flight process and the control simulation with failures.
Aimed at the complex flight control system of a fighter, a kind of robust adapti ve control methods using multiple models is presented to make the control system track the given signal under different working conditions and to reconfigure th e control law for some structural failures. Firstly, the multiplemodel control structure is formed by several linear models and one fuzzy model. In the fuzzy logic way, weights of the multiplemodel adaptive controller are obtained. Then , a dynamic structure adaptive neural network is introduced to stabilize the who le system and eliminate the influence caused by the frequent switching. Simulati on results show that the control method is effective by demonstrating the normal flight process and the control simulation with failures.
2010(3):226-231.
Abstract:
By combining the distributed Kalman filter (DKF) with the back propagation neural network (BPNN), a novel method is proposed to identify the bias of electrostatic suspended gyroscope (ESG). Firstl y, the data sets of multimeasurements of the same ESG in different noise env ironments are ＂mapped＂ into a sensor network, and DKF with embedded consensus filters is then used to preprocess the data sets. After transforming the preproc essed results into the trained input and the desired output of neural network, B PNN with the learning rate and the momentum term is further utilized to identify the ESG bias. As demonstrated in the experiment,the proposed approach is effect ive for the model identification of the ESG bias.
By combining the distributed Kalman filter (DKF) with the back propagation neural network (BPNN), a novel method is proposed to identify the bias of electrostatic suspended gyroscope (ESG). Firstl y, the data sets of multimeasurements of the same ESG in different noise env ironments are ＂mapped＂ into a sensor network, and DKF with embedded consensus filters is then used to preprocess the data sets. After transforming the preproc essed results into the trained input and the desired output of neural network, B PNN with the learning rate and the momentum term is further utilized to identify the ESG bias. As demonstrated in the experiment,the proposed approach is effect ive for the model identification of the ESG bias.
2010(3):232-238.
Abstract:
A new kind of composite fillers, composed of AgCuTi alloy and nanoTi C powders, is utilized to braze cubic boron nitride (CBN) grains and tool substr ate. The bonding system, including the interfacial microstructure and reactive p roducts between CBN grains and filler layer, is observed by optical microscope a nd scanning electron microscope (SEM). Resistanttowear experiments of the br azed grains are performed. Results show that the nanoTiC powders evenly di stribute in the filler layer so that the resultants grow compactly and uniformly on the surface of CBN grain. This indicates that the chemical bond is established between CBN grains and nanoTiC modified filler. Accordingly, the bonding strength of the g rains is ensured. The CBN grains are worn smoothly without grain pullout.
A new kind of composite fillers, composed of AgCuTi alloy and nanoTi C powders, is utilized to braze cubic boron nitride (CBN) grains and tool substr ate. The bonding system, including the interfacial microstructure and reactive p roducts between CBN grains and filler layer, is observed by optical microscope a nd scanning electron microscope (SEM). Resistanttowear experiments of the br azed grains are performed. Results show that the nanoTiC powders evenly di stribute in the filler layer so that the resultants grow compactly and uniformly on the surface of CBN grain. This indicates that the chemical bond is established between CBN grains and nanoTiC modified filler. Accordingly, the bonding strength of the g rains is ensured. The CBN grains are worn smoothly without grain pullout.
2010(3):239-247.
Abstract:
geometric modeling method for generatingmanufactured spiral bevel gears(SBGs ) is proposed. It consists of two steps: (1) creating a reference model by simula ting the process of cutting spiral bevel gear, (2) reconstructing the final soli d model by collecting data points from the reference model and fitting these poi nts into NURBS surfaces. In this method, cutting simulation avoids abstruse math ematical theories and complex methods, thus making it convenient to obtain data points o n the complex tooth surface before the gear is manufactured and efficient to inc rease the accuracy of the solid model. Also, the representations of tooth surfac es of the final model is unified as a NURBS surface function. The NURBS surface is continuous and smooth, thus it is available for wide applica tions i n CAD／CAE. The experiment proves that the method can be used to establish an a ccura te pair of SBG models, thus providing a feasible and effective way for CAD／CAE modeling.
geometric modeling method for generatingmanufactured spiral bevel gears(SBGs ) is proposed. It consists of two steps: (1) creating a reference model by simula ting the process of cutting spiral bevel gear, (2) reconstructing the final soli d model by collecting data points from the reference model and fitting these poi nts into NURBS surfaces. In this method, cutting simulation avoids abstruse math ematical theories and complex methods, thus making it convenient to obtain data points o n the complex tooth surface before the gear is manufactured and efficient to inc rease the accuracy of the solid model. Also, the representations of tooth surfac es of the final model is unified as a NURBS surface function. The NURBS surface is continuous and smooth, thus it is available for wide applica tions i n CAD／CAE. The experiment proves that the method can be used to establish an a ccura te pair of SBG models, thus providing a feasible and effective way for CAD／CAE modeling.
2010(3):248-254.
Abstract:
Warm tensile tests for aluminum alloy 7022 sheet are held at different temperatures and strain rates. The range of temperature is 293－773 K and that of the strain rate is 0. 001－01 s-1. The warm tensile properties, relations am on g temperature, strain rate, and the flow stress are discussed. Constitutive equa t ions under the warm tension are obtained based on revised Hooke law and Grosman equa tion. It is concluded that flow stress declines with the increase of the tempera ture and the decrease of the strain rates. The elongation percentage increases with the increase of the temperature and the decrease of strain rate.
Warm tensile tests for aluminum alloy 7022 sheet are held at different temperatures and strain rates. The range of temperature is 293－773 K and that of the strain rate is 0. 001－01 s-1. The warm tensile properties, relations am on g temperature, strain rate, and the flow stress are discussed. Constitutive equa t ions under the warm tension are obtained based on revised Hooke law and Grosman equa tion. It is concluded that flow stress declines with the increase of the tempera ture and the decrease of the strain rates. The elongation percentage increases with the increase of the temperature and the decrease of strain rate.
2010(3):255-260.
Abstract:
Bagging is not quite suitable for stable classifiers such as nearest n eighbor classifiers due to the lack of diversity and it is difficult to be direc tly applied to face recognition as well due to the small sample size (SSS) prope rty of face recognition. To solve the two problems, local Bagging (LBagging) i s proposed to simultaneously make Bagging apply to both nearest neighbor classifi ers and face recognition. The major difference between LBagging and Bagging is that LBagging performs the bootstrap sampling on each local region partitioned from the original face image rather than the whole face image. Since the dimensi onality of local region is usually far less than the number of samples and the c omponent classifiers are constructed just in different local regions, LBagging deals with SSS problem and generates more diverse component c lassifiers. Experimental results on four standard face image databases (AR, Yale , ORL and Yale B) indicate that the proposed LBagging method is effective and robust to llumination, occlusion and slight pose variation.
Bagging is not quite suitable for stable classifiers such as nearest n eighbor classifiers due to the lack of diversity and it is difficult to be direc tly applied to face recognition as well due to the small sample size (SSS) prope rty of face recognition. To solve the two problems, local Bagging (LBagging) i s proposed to simultaneously make Bagging apply to both nearest neighbor classifi ers and face recognition. The major difference between LBagging and Bagging is that LBagging performs the bootstrap sampling on each local region partitioned from the original face image rather than the whole face image. Since the dimensi onality of local region is usually far less than the number of samples and the c omponent classifiers are constructed just in different local regions, LBagging deals with SSS problem and generates more diverse component c lassifiers. Experimental results on four standard face image databases (AR, Yale , ORL and Yale B) indicate that the proposed LBagging method is effective and robust to llumination, occlusion and slight pose variation.
2010(3):261-268.
Abstract:
Locality preserving projection (LPP) is a typical and popular dimensi onality reduction (DR) method, and it can potentially find discriminative projec tion directions by preserving the local geometric structure in data. However, LP P is based on the neighborhood graph artificially constructed from the original data , and the performance of LPP relies on how well the neares t ne ighbor criterion work in the original space. To address this issue, a novel DR algo rithm, called the selfdependent LPP (sdLPP) is proposed. And it is based on th e fact that the nearest neighbor criterion usually achieves better performance in LPP t ransformed space than that in the original space. Firstly, LPP is performed base d on the typical neighborhood graph; then, a new neighborhood graph is constructed i n LPP transformed space and repeats LPP. Furthermore, a new criterion, called the impro ved Laplacian score, is developed as an empirical reference for the discriminati v e power and the iterative termination. Finally, the feasibility and the effectiv enes s of the method are verified by several publicly available UCI and face data set s with promising results.
Locality preserving projection (LPP) is a typical and popular dimensi onality reduction (DR) method, and it can potentially find discriminative projec tion directions by preserving the local geometric structure in data. However, LP P is based on the neighborhood graph artificially constructed from the original data , and the performance of LPP relies on how well the neares t ne ighbor criterion work in the original space. To address this issue, a novel DR algo rithm, called the selfdependent LPP (sdLPP) is proposed. And it is based on th e fact that the nearest neighbor criterion usually achieves better performance in LPP t ransformed space than that in the original space. Firstly, LPP is performed base d on the typical neighborhood graph; then, a new neighborhood graph is constructed i n LPP transformed space and repeats LPP. Furthermore, a new criterion, called the impro ved Laplacian score, is developed as an empirical reference for the discriminati v e power and the iterative termination. Finally, the feasibility and the effectiv enes s of the method are verified by several publicly available UCI and face data set s with promising results.
2010(3):269-274.
Abstract:
A new method for the white blood cell (WBC) detection is presented based on the relevance vector machine (RVM). Firstly, the sparse relevance vectors (RVs) are obtained while fitting the 1D histogram by RVM. Then, the needed threshold val ue is directly selected from these limited RVs. Finally, the entire connective W BC regions are segmented from the original image. The method is used for the WBC detection. It reduces the interference induced by the illumination and the stain ing. It has advantages of the high computation efficiency and the no extra param eter setting. Experimental results demonstrate good performances of the method.
A new method for the white blood cell (WBC) detection is presented based on the relevance vector machine (RVM). Firstly, the sparse relevance vectors (RVs) are obtained while fitting the 1D histogram by RVM. Then, the needed threshold val ue is directly selected from these limited RVs. Finally, the entire connective W BC regions are segmented from the original image. The method is used for the WBC detection. It reduces the interference induced by the illumination and the stain ing. It has advantages of the high computation efficiency and the no extra param eter setting. Experimental results demonstrate good performances of the method.
2010(3):275-279.
Abstract:
A compact fourcomponent twodimen sional (2-D） finitedifference frequency domain (FDFD) method with the equiv alent sur face impedance boundary condition is used to analyze the dispersion characte ristics of multilayer metalcoated waveguides. According to the equivalent surf ace impedance boundary condition, the relationship between transverse field comp onents on the boundary can be easily depicted. Once the eigen equation is solve d, the propagation constant can be obtained as the eigen value for a given frequ ency. Results of the proposed method agaree well with those of high frequency st ructure simulator(HFSS).
A compact fourcomponent twodimen sional (2-D） finitedifference frequency domain (FDFD) method with the equiv alent sur face impedance boundary condition is used to analyze the dispersion characte ristics of multilayer metalcoated waveguides. According to the equivalent surf ace impedance boundary condition, the relationship between transverse field comp onents on the boundary can be easily depicted. Once the eigen equation is solve d, the propagation constant can be obtained as the eigen value for a given frequ ency. Results of the proposed method agaree well with those of high frequency st ructure simulator(HFSS).
CHINA AERO-SCIENCES
Mobile website
Transactions of Nanjing University of Aeronautics & Astronautics ® 2024 All Rights Reserved