Mechanism Design with MP-Neural Networks

1998 ◽  
Vol 120 (4) ◽  
pp. 527-532 ◽  
Author(s):  
J. Li ◽  
K. C. Gupta
Keyword(s):  
Error Function ◽  
Least Square ◽  
Numerical Experience ◽  
Variable Approach ◽  
Structural Error ◽  
Augmented Lagrange Multiplier ◽  
Equation Error ◽  
Complex Structural ◽  
Alm Method ◽  
New Algorithms

The prevalent Mathematical Programming Neural Network (MPNN) models are surveyed, and MPNN models have been developed and applied to the unconstrained optimization of mechanisms. Algorithms which require Hessian inversion and those which build up a variable approach matrix, are investigated. Based upon a comprehensive investigation of the Augmented Lagrange Multiplier (ALM) method, new algorithms have been developed from the combination of ideas from MPNN and ALM methods and applied to the constrained optimization of mechanisms. A relationship between the weighted least square minimization of design equation error residuals and the mini-max norm of the structure error for function generating mechanisms is developed and employed in the optimization process; as a result, the computational difficulties arising from the direct usage of the complex structural error function have been avoided. The paper presents relevant theory as well as some numerical experience for four MPNN algorithms.

Mathematical Programming Neural Networks (MPNN) for Mechanism Design

1997 ◽  
Author(s):  
Jianmin Li ◽  
Krishna C. Gupta
Keyword(s):  
Mathematical Programming ◽  
Error Function ◽  
Least Square ◽  
Numerical Experience ◽  
Variable Approach ◽  
Structural Error ◽  
Augmented Lagrange Multiplier ◽  
Complex Structural ◽  
Alm Method ◽  
New Algorithms

Abstract The prevalent Mathematical Programming Neural Network (MPNN) models are surveyed, and MPNN models have been developed and applied to the unconstrained optimization of mechanisms. Algorithms which require Hessian inversion and those which build up a variable approach matrix, are investigated. Based upon a comprehensive investigation of the Augmented Lagrange Multiplier (ALM) method, new algorithms have been developed from the combination of ideas from MPNN and ALM methods and applied to the constrained optimization of mechanisms. A relationship between the weighted least square minimization of design equation error residuals and the mini-max norm of the structure error for function generating mechanisms is developed and employed in the optimization process; as a result, the computational difficulties arising from the direct usage of the complex structural error function have been avoided. The paper presents relevant theory as well as some numerical experience for four MPNN algorithms.

A More Effective Formulation of the Path Generation Mechanism Synthesis Problem

1994 ◽  
Author(s):  
Irfan Ullah ◽  
Sridhar Kota
Keyword(s):  
Objective Function ◽  
Error Function ◽  
Mathematical Optimization ◽  
Optimization Methods ◽  
Complex Nature ◽  
Mechanism Synthesis ◽  
Structural Error ◽  
Crank Angle ◽  
The Common ◽  
Selection Of

Abstract Use of mathematical optimization methods for synthesis of path-generating mechanisms has had only limited success due to the very complex nature of the commonly used Structural Error objective function. The complexity arises, in part, because the objective function represents not only the error in the shape of the coupler curve, but also the error in location, orientation and size of the curve. Furthermore, the common introduction of timing (or crank angle), done generally to facilitate selection of corresponding points on the curve for calculating structural error, has little practical value and unnecessarily limits possible solutions. This paper proposes a new objective function, based on Fourier Descriptors, which allows search for coupler curve of the desired shape without reference to location, orientation, or size. The proposed objective function compares overall shape properties of curves rather than making point-by-point comparison and therefore does not requires prescription of timing. Experimental evidence is provided to show that it is much easier to search the space of the proposed objective function compared to the structural error function.

An Improved Object-Based Video Segmentation Method

2012 ◽  
Vol 220-223 ◽  
pp. 2445-2449
Author(s):  
Wen Dan Xu ◽  
Xin Quan Lai ◽  
Dong Lai Xu
Keyword(s):  
Motion Estimation ◽  
Video Compression ◽  
High Performance ◽  
Video Segmentation ◽  
Error Function ◽  
Least Square ◽  
Video Frame ◽  
Merging Process ◽  
Initial Segmentation ◽  
Two Stages

This paper presents an improved video segmentation scheme, which consists of two stages: initial segmentation and motion estimation. In the initial segmentation, the watershed transformation followed by a region adjacency graph guided region merging process is used to partition the first video frame into spatial homogenous regions. Then the motion of changed region is estimated. Based on the highly efficient quadratic motion model, the motion estimation is undertaken using Gauss-Newton Levenberg-Marquardt method to minimize the least-square error function. Experimental results show the proposed scheme provides high performance in terms of segmentation accuracy and video compression ratio.

scholarly journals JointL1/2-Norm Constraint and Graph-Laplacian PCA Method for Feature Extraction

2017 ◽  
Vol 2017 ◽  
pp. 1-14 ◽  
Author(s):  
Chun-Mei Feng ◽  
Ying-Lian Gao ◽  
Jin-Xing Liu ◽  
Juan Wang ◽  
Dong-Qin Wang ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Error Function ◽  
Principal Component ◽  
Graph Laplacian ◽  
Specific Gene ◽  
Data Sets ◽  
Pca Method ◽  
Norm Constraint ◽  
Alm Method ◽  
Pca Algorithm

Principal Component Analysis (PCA) as a tool for dimensionality reduction is widely used in many areas. In the area of bioinformatics, each involved variable corresponds to a specific gene. In order to improve the robustness of PCA-based method, this paper proposes a novel graph-Laplacian PCA algorithm by adoptingL1/2constraint (L1/2gLPCA) on error function for feature (gene) extraction. The error function based onL1/2-norm helps to reduce the influence of outliers and noise. Augmented Lagrange Multipliers (ALM) method is applied to solve the subproblem. This method gets better results in feature extraction than other state-of-the-art PCA-based methods. Extensive experimental results on simulation data and gene expression data sets demonstrate that our method can get higher identification accuracies than others.

Least Square Inversed Analysis of Soil Parameter for Foundation with Two-Order Gradient Theoretic Method

2011 ◽  
Vol 243-249 ◽  
pp. 2294-2299 ◽  
Author(s):  
Yao Feng Xie
Keyword(s):  
Differential Equations ◽  
Error Function ◽  
Least Square Method ◽  
Least Square ◽  
Gradient Theory ◽  
Mechanical Theory ◽  
Soil Parameter ◽  
True Value ◽  
Initial Soil ◽  
Calculation Results

Combined with two-order gradient theory, the least square inversed analysis of the soil parameter for the foundation is studied and put forward in detail. After the mechanical theory for the plate on the foundation is introduced, the controlling differential equations of the plate on the foundation which is subjected to vertical loads are deduced. Through utilizing Fourier transformative theory, the corresponding solutions to the plate on the foundation are gained. Linear algebra controlling equations for the plate are achieved which leads to solve the original differential equations more easily. The least square error function for the soil parameter on the plate is established and applied with the two-order gradient method. The inversed steps on the least square error function for the soil parameter are listed. The calculation results verify the conclusions that the soil parameter of the foundation can be efficiently inversed by applying the least square theory. When different initial soil parameter is set, the iterative computations can be convergent to the true value of the soil parameter. And this least square method can also be applied for the problem of inversed analysis of parameters for other foundation models.

On the transmission efficiency of a spring-type operating kinematic mechanism for an SF6 gas-insulated circuit breaker

2001 ◽  
Vol 215 (10) ◽  
pp. 1239-1249 ◽  
Author(s):  
F-C Chen ◽  
M-H Hsu
Keyword(s):  
Degrees Of Freedom ◽  
Circuit Breaker ◽  
Transmission Efficiency ◽  
Cam Mechanism ◽  
Augmented Lagrange Multiplier ◽  
Transmission Angle ◽  
Present Design ◽  
Spring Type ◽  
Alm Method ◽  
Kinematic Mechanism

The purpose of this paper is to study the transmission efficiency of a spring-type operating kinematic mechanism for an SF6 gas-insulated circuit breaker. This mechanism has two degrees of freedom, which, when operating in the open, close, or return mode, becomes a single degree of freedom by fixing a link that adjoins the frame to achieve specific requirements. Firstly, the vector-loop method is employed for position analysis of the links of the mechanism. Subsequently, by using coordinate transformation and the geometry of the cam and the follower, the relationships between cam rotation and the position of the linkage follower in the close operation, as well as cam rotation and the position of the motor-switch follower are established. Based on the transmission angle of the linkage and the pressure angle of the cam mechanism, the transmission efficiencies of the mechanism are evaluated and discussed. The mechanism is also computer simulated and the transmission angle is optimized using the augmented Lagrange multiplier (ALM) method. The results show that the optimized parameters noticeably improve the transmission angles of the present design.

On the Parameter Identification of LuGre Tire Friction Model

2013 ◽  
Author(s):  
Yusuke Matsutani ◽  
Hiroyuki Sugiyama
Keyword(s):  
Error Function ◽  
Numerical Procedure ◽  
Friction Model ◽  
Least Square ◽  
Model Parameters ◽  
Force Model ◽  
Lugre Model ◽  
Tire Force ◽  
Brake Force ◽  
Tire Friction

In this investigation, use of the LuGre tire friction model for the transient brake force analysis is discussed. In particular, a numerical procedure for estimating parameters for the LuGre tire force model is developed. The parameters of the distributed LuGre model are identified such that the error function of tire forces obtained using the model and experiment can be minimized. Friction parameters used in the LuGre tire force model are estimated using the characteristics curve of the friction coefficient as a function of the slip velocity first, and then the adhesion parameter is estimated using the slope around the zero slip ratio using the least square fitting. Iterative solution procedures are then employed to identify the overall model parameters using the initial estimates provided. It is demonstrated that use of the proposed numerical procedure leads to accurate prediction of the LuGre model parameters for various loading and speed conditions. Furthermore, it is demonstrated that the decrease in the peak of the brake force as the increase in the running speed can be well predicted using the transient distributed LuGre tire force model with model parameters predicated using the proposed procedure.

Identification and Compensation of a Robot Kinematic Parameter for Positioning Accuracy Improvement

Robotica ◽  
1991 ◽  
Vol 9 (1) ◽  
pp. 99-105 ◽  
Author(s):  
D. H. Kim ◽  
K. H. Cook ◽  
J. H. Oh
Keyword(s):  
Error Function ◽  
Least Square Method ◽  
Kinematic Parameter ◽  
Least Square ◽  
Joint Angles ◽  
Kinematic Parameters ◽  
Accuracy Improvement ◽  
Coordinate Frames ◽  
Identification Method ◽  
Value Decomposition

SUMMARYThis paper presents a simple identification method of the actual kinematic parameters for a robot with parallel joints. It is known that Denavit–Hartenberg's coordinate System is not useful for nearly parallel joints. In this paper, the coordinate frames are reassigned to model the kinematic parameter between nearly parallel joints by four parameters. The proposed identification method uses a straight ruler about 1 m long. A robot hand is placed by using a teaching pendant at the prescribed points on the ruler, and the corresponding error function is defined. The identified kinematic parameters, which make the error function zero, are obtained by the iterative least square method based on the singular value decomposition. In the compensation of joint angles, only the position is considered because the usual applications of robot do not require a precise orientation control.

Evaluating the total factor productivity spillovers from FDI: Experience of South Asia

2021 ◽  
Author(s):  
Sanjoy Kumar Saha ◽  
Se Eun Jeong
Keyword(s):  
Growth Rate ◽  
South Asia ◽  
Jel Classification ◽  
Absorption Capacity ◽  
Least Square ◽  
Malmquist Productivity Index ◽  
Productivity Index ◽  
Factor Productivity ◽  
Productivity Spillovers ◽  
Variable Approach

Abstract South Asia is a region of impressive growth rate over the years. Factor productivity (TFP) is also found to be notable. Thus, the relevant question is how the TFP is boosted up which held the growth rate up for such a long time. Foreign direct investment(FDI) can be the one source that may generate the productivity spillovers. As a result, the motive of the paper is to find out the role or contribution of FDI in the productivity spillovers in five South Asian countries. Deriving TFP from Malmquist productivity index we found out that overall contribution of FDI in productivity spillover is very negligible or scanty. We address the issue of cross country time invariant heterogeneity through fixed effect and least square dummy variable (LSDV) method and endogeneity problem through instrumental variable approach. We observe that the scarcity of absorption capacity measured by human capital is the major obstacle for not having any worth mentioning role of FDI in TFP. Therefore, our study highlights on the importance to improve the quality of higher education as well as expanding expenditure on educational purpose which is a matter of serious concern in this region.JEL classification: F6, F14, F62, F21, F23, O33

Sign in / Sign up

Export Citation Format

Share Document