scholarly journals A Novel Measurement Matrix Optimization Approach for Hyperspectral Unmixing

2017 ◽  
Vol 2017 ◽  
pp. 1-13
Author(s):  
Su Xu ◽  
Xiping He
Keyword(s):  
Computational Cost ◽  
Descent Method ◽  
Qr Factorization ◽  
Eigenvalue Decomposition ◽  
Optimization Approach ◽  
Gradient Descent Method ◽  
Measurement Matrix ◽  
Hyperspectral Unmixing ◽  
Matrix Optimization ◽  
Reconstruction Performance

Each pixel in the hyperspectral unmixing process is modeled as a linear combination of endmembers, which can be expressed in the form of linear combinations of a number of pure spectral signatures that are known in advance. However, the limitation of Gaussian random variables on its computational complexity or sparsity affects the efficiency and accuracy. This paper proposes a novel approach for the optimization of measurement matrix in compressive sensing (CS) theory for hyperspectral unmixing. Firstly, a new Toeplitz-structured chaotic measurement matrix (TSCMM) is formed by pseudo-random chaotic elements, which can be implemented by a simple hardware; secondly, rank revealing QR factorization with eigenvalue decomposition is presented to speed up the measurement time; finally, orthogonal gradient descent method for measurement matrix optimization is used to achieve optimal incoherence. Experimental results demonstrate that the proposed approach can lead to better CS reconstruction performance with low extra computational cost in hyperspectral unmixing.

Nonlinear System Identification Using a Neo Fuzzy Neuron Algorithm: Electrical Drive Application

1999 ◽  
Vol 09 (03) ◽  
pp. 211-217 ◽  
Author(s):  
RÉGIS P. LANDIM ◽  
BENJAMIM R. DE MENEZES ◽  
SELÊNIO R. SILVA ◽  
WALMIR M. CAMINHAS
Keyword(s):  
Nonlinear System Identification ◽  
Computational Cost ◽  
Descent Method ◽  
Point Of View ◽  
Gradient Descent Method ◽  
Flux Observer ◽  
Rotor Flux ◽  
On Line ◽  
One Step ◽  
Motor Drive System

This work presents a Neo-Fuzzy-Neuron algorithm for the identification of nonlinear dynamic systems at the point of view of a rotor flux observer. The algorithm training is on-line, has low computational cost, does not require previous training and its convergence in one step is proved. The gradient descent method is used for its weights adjustment. Simulation and experimental results demonstrate the effetiveness of the algorithm for flux observer of induction motor drive system.

scholarly journals A New Method of Measurement Matrix Optimization for Compressed Sensing Based on Alternating Minimization

Mathematics ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. 329
Author(s):  
Renjie Yi ◽  
Chen Cui ◽  
Biao Wu ◽  
Yang Gong
Keyword(s):  
Compressed Sensing ◽  
New Method ◽  
Gram Matrix ◽  
Alternating Minimization ◽  
Measurement Matrix ◽  
Mutual Coherence ◽  
Matrix Optimization ◽  
Shrinkage Function ◽  
Reconstruction Performance ◽  
Method Of Measurement

In this paper, a new method of measurement matrix optimization for compressed sensing based on alternating minimization is introduced. The optimal measurement matrix is formulated in terms of minimizing the Frobenius norm of the difference between the Gram matrix of sensing matrix and the target one. The method considers the simultaneous minimization of the mutual coherence indexes including maximum mutual coherence μmax, t-averaged mutual coherence μave and global mutual coherence μall, and solves the problem that minimizing a single index usually results in the deterioration of the others. Firstly, the threshold of the shrinkage function is raised to be higher than the Welch bound and the relaxed Equiangular Tight Frame obtained by applying the new function to the Gram matrix is taken as the initial target Gram matrix, which reduces μave and solves the problem that μmax would be larger caused by the lower threshold in the known shrinkage function. Then a new target Gram matrix is obtained by sequentially applying rank reduction and eigenvalue averaging to the initial one, leading to lower. The analytical solutions of measurement matrix are derived by SVD and an alternating scheme is adopted in the method. Simulation results show that the proposed method simultaneously reduces the above three indexes and outperforms the known algorithms in terms of reconstruction performance.

scholarly journals Mesoscale Modelling of Concretes Subjected to Triaxial Loadings: Mechanical Properties and Fracture Behaviour

Materials ◽  
2021 ◽  
Vol 14 (5) ◽  
pp. 1099
Author(s):  
Qingqing Chen ◽  
Yuhang Zhang ◽  
Tingting Zhao ◽  
Zhiyong Wang ◽  
Zhihua Wang
Keyword(s):  
Mechanical Properties ◽  
Tensile Stress ◽  
Failure Criterion ◽  
Mesoscale Model ◽  
Fracture Behaviour ◽  
Descent Method ◽  
Triaxial Stress ◽  
Gradient Descent Method ◽  
Stress States ◽  
Shear Failures

The mechanical properties and fracture behaviour of concretes under different triaxial stress states were investigated based on a 3D mesoscale model. The quasistatic triaxial loadings, namely, compression–compression–compression (C–C–C), compression–tension–tension (C–T–T) and compression–compression–tension (C–C–T), were simulated using an implicit solver. The mesoscopic modelling with good robustness gave reliable and detailed damage evolution processes under different triaxial stress states. The lateral tensile stress significantly influenced the multiaxial mechanical behaviour of the concretes, accelerating the concrete failure. With low lateral pressures or tensile stress, axial cleavage was the main failure mode of the specimens. Furthermore, the concretes presented shear failures under medium lateral pressures. The concretes experienced a transition from brittle fracture to plastic failure under high lateral pressures. The Ottosen parameters were modified by the gradient descent method and then the failure criterion of the concretes in the principal stress space was given. The failure criterion could describe the strength characteristics of concrete materials well by being fitted with experimental data under different triaxial stress states.

Dynamic Optimization of Steel Catenary Risers Based on Reliability Using Metamodel

2010 ◽  
Author(s):  
Wenqing Zheng ◽  
Hezhen Yang
Keyword(s):  
Dynamic Optimization ◽  
Computational Cost ◽  
Minimum Cost ◽  
Probabilistic Constraints ◽  
Optimization Approach ◽  
Element Analysis ◽  
Steel Catenary Riser ◽  
Reliability Based Design ◽  
Minimum Cost Design ◽  
Steel Catenary Risers

Reliability based design optimization (RBDO) of a steel catenary riser (SCR) using metamodel is investigated. The purpose of the optimization is to find the minimum-cost design subjecting to probabilistic constraints. To reduce the computational cost of the traditional double-loop RBDO, a single-loop RBDO approach is employed. The performance function is approximated by using metamodel to avoid time consuming finite element analysis during the dynamic optimization. The metamodel is constructed though design of experiments (DOE) sampling. In addition, the reliability assessment is carried out by Monte Carlo simulations. The result shows that the RBDO of SCR is a more rational optimization approach compared with traditional deterministic optimization, and using metamodel technique during the dynamic optimization process can significantly decrease the computational expense without sacrificing accuracy.

scholarly journals Fuzzy Formation Control and Collision Avoidance for Multiagent Systems

2013 ◽  
Vol 2013 ◽  
pp. 1-18 ◽  
Author(s):  
Yeong-Hwa Chang ◽  
Chun-Lin Chen ◽  
Wei-Shou Chan ◽  
Hung-Wei Lin ◽  
Chia-Wen Chang
Keyword(s):  
Collision Avoidance ◽  
Multiagent Systems ◽  
Formation Control ◽  
Descent Method ◽  
Minimum Problem ◽  
Gradient Descent Method ◽  
Consensus Formation ◽  
Distributed Information ◽  
Graph Theoretic ◽  
Learning Rules

This paper aims to investigate the formation control of leader-follower multiagent systems, where the problem of collision avoidance is considered. Based on the graph-theoretic concepts and locally distributed information, a neural fuzzy formation controller is designed with the capability of online learning. The learning rules of controller parameters can be derived from the gradient descent method. To avoid collisions between neighboring agents, a fuzzy separation controller is proposed such that the local minimum problem can be solved. In order to highlight the advantages of this fuzzy logic based collision-free formation control, both of the static and dynamic leaders are discussed for performance comparisons. Simulation results indicate that the proposed fuzzy formation and separation control can provide better formation responses compared to conventional consensus formation and potential-based collision-avoidance algorithms.

Sign in / Sign up

Export Citation Format

Share Document