Topology Optimization for Piecewise Dynamic Models Using a Genetic Algorithm

2010 ◽  
Author(s):  
Yi Liu ◽  
Dragan Djurdjanovic
Keyword(s):  
Genetic Algorithm ◽  
Topology Optimization ◽  
Dynamic Model ◽  
Dynamic Models ◽  
Optimization Procedure ◽  
Nonlinear Dynamic System ◽  
Local Model ◽  
Model Parameters ◽  
Highly Nonlinear ◽  
Node Connectivity

It has been demonstrated in the previous research that the node connectivity in the graph encoding the topological neighborhood relationships between local models in a piecewise dynamic model may significantly affect the cooperative learning process. It was shown that a graph with a larger connectivity leads to a quicker learning adaption due to more rapidly decaying transients of the estimation of local model parameters. In the same time, it was shown that the accuracy could be degraded by a larger bias in the asymptotic portion of the estimations of local model parameters. The efforts in topology optimization should therefore strive towards a high accuracy of the asymptotic portion of the estimator of local model parameters while simultaneously accelerating the decay of the estimation transients. In this paper, we pursue minimization of the residual sum of squares of a piecewise dynamic model after a predetermined number of training steps. The optimization of inter-model topology is implemented via a genetic algorithm that manipulates adjacency matrices of the graph underlying the piecewise dynamic model. An example of applying the topology optimization procedure on a peicewise linear model of a highly nonlinear dynamic system is provided to show the efficacy of the new method.

Sensor Dynamic Modeling Based on LS-SVM and NGA

2008 ◽  
Vol 381-382 ◽  
pp. 439-442
Author(s):  
Qi Wang ◽  
Zhi Gang Feng ◽  
K. Shida
Keyword(s):  
Genetic Algorithm ◽  
Neural Networks ◽  
Support Vector Machine ◽  
Least Squares ◽  
Dynamic Model ◽  
Dynamic Modeling ◽  
Support Vector ◽  
Local Minima ◽  
Highly Nonlinear ◽  
Sharing Function

Least squares support vector machine (LS-SVM) combined with niche genetic algorithm (NGA) are proposed for nonlinear sensor dynamic modeling. Compared with neural networks, the LS-SVM can overcome the shortcomings of local minima and over fitting, and has higher generalization performance. The sharing function based niche genetic algorithm is used to select the LS-SVM parameters automatically. The effectiveness and reliability of this method are demonstrated in two examples. The results show that this approach can escape from the blindness of man-made choice of LS-SVM parameters. It is still effective even if the sensor dynamic model is highly nonlinear.

A parameter identification method for robot dynamic models using a balancing mechanism

Robotica ◽  
1989 ◽  
Vol 7 (4) ◽  
pp. 327-337 ◽  
Author(s):  
T. G. Lim ◽  
H. S. Cho ◽  
W. K. Chung
Keyword(s):  
Dynamic Model ◽  
Dynamic Models ◽  
Least Square ◽  
Estimation Accuracy ◽  
Model Parameters ◽  
Robot Dynamics ◽  
Dynamic Parameters ◽  
Joint Torques ◽  
Motion Data ◽  
Identification Method

SUMMARYAccurate modeling of robot dynamics is a prerequisite for the design of model-based control schemes and enhancement of the performance of the robot. The dynamic parameters associated with a pseudo-inertia matrix are often difficult to identify accurately because the inertia torques are small in comparison to gravity loadings, thus creating signal processing problem. The identification method presented in this paper utilizes a balancing mechanism which increases the estimation accuracy of the dynamic parameters. The balancing mechanism has the effect of amplifying the inertia-related torque signal by eliminating gravity loadings acting on the robot joints. A series of motion data were experimentally obtained through sequential test steps. By incorporating the measured information about joint torques, angular positions, velocities and accelerations the least square algorithm was used to identify the dynamic parameters. The estimated values were converted to those of the original robot model to obtain its dynamic model parameters. The identified robot dynamic model was shown to be accurate enough to predict the actual robot motions.

Nonlinear Dynamic Model of a Two-Stage Pressure Relief Valve for Designers

2001 ◽  
Vol 124 (1) ◽  
pp. 62-66 ◽  
Author(s):  
Pei-Sun Zung ◽  
Ming-Hwei Perng
Keyword(s):  
Dynamic Model ◽  
Nonlinear Dynamic ◽  
Dynamic Models ◽  
Operating Conditions ◽  
Model Parameters ◽  
Nonlinear Dynamic Model ◽  
Relief Valve ◽  
Two Stage ◽  
Pressure Relief ◽  
Wide Range

This paper presents a handy nonlinear dynamic model for the design of a two stage pilot pressure relief servo-valve. Previous surveys indicate that the performance of existing control valves has been limited by the lack of an accurate dynamic model. However, most of the existing dynamic models of pressure relief valves are developed for the selection of a suitable valve for a hydraulic system, and assume model parameters which are not directly controllable during the manufacturing process. As a result, such models are less useful for a manufacturer eager to improve the performance of a pressure valve. In contrast, model parameters in the present approach have been limited to dimensions measurable from the blue prints of the valve such that a specific design can be evaluated by simulation before actually manufacturing the valve. Moreover, the resultant model shows excellent agreement with experiments in a wide range of operating conditions.

scholarly journals Multi-Objective Optimization of Experiments Using Curvature and Fisher Information Matrix

2017 ◽  
Author(s):  
Erica Manesso ◽  
Sridharan Srinath ◽  
Rudiyanto Gunawan
Keyword(s):  
Fisher Information ◽  
Fisher Information Matrix ◽  
Dynamic Models ◽  
Information Matrix ◽  
Model Parameters ◽  
Multi Objective Optimization ◽  
Experimental Conditions ◽  
Maximum Information ◽  
Multi Objective ◽  
Highly Nonlinear

The bottleneck in creating dynamic models of biological networks and processes often lies in estimating unknown kinetic model parameters from experimental data. In this regard, experimental conditions have a strong influence on parameter identifiability and should therefore be optimized to give the maximum information for parameter estimation. Existing model-based design of experiment (MBDOE) methods commonly rely on the Fisher Information Matrix (FIM) for defining a metric of data informativeness. When the model behavior is highly nonlinear, FIM-based criteria may lead to suboptimal designs since the FIM only accounts for the linear variation of the model outputs with respect to the parameters. In this work, we developed a multi-objective optimization (MOO) MBDOE, where model nonlinearity was taken into consideration through the use of curvature. The proposed MOO MBDOE involved maximizing data informativeness using a FIM-based metric and at the same time minimizing the model curvature. We demonstrated the advantages of the MOO MBDOE over existing FIM-based and other curvature-based MBDOEs in an application to the kinetic modeling of fed-batch fermentation of Baker's yeast.

Linear parameter variant modeling and parameter identification of a cable-driven micromanipulator for surgical robot

2018 ◽  
Vol 233 (5) ◽  
pp. 1828-1840 ◽  
Author(s):  
Wenjie Wang ◽  
Lingtao Yu ◽  
Jing Yang
Keyword(s):  
Genetic Algorithm ◽  
Dynamic Model ◽  
Joint Angle ◽  
Linear Time ◽  
Surgical Robot ◽  
Least Square ◽  
Feedback Signal ◽  
Model Parameters ◽  
Linear Parameter ◽  
Nonlinear Dynamic Model

This paper proposes a novel cable-driven micromanipulator for surgical robots. A single-joint principle prototype for surgical robot micromanipulator was manufactured to test the proposed design. Elasticity and friction were assessed to establish a joint angle estimator; estimator parameters were obtained by a combination of least square method and genetic algorithm. Angle closed-loop control was performed by considering the joint angle estimator output as the feedback signal. A nonlinear dynamic model was established in the state-space and described as a linear parameter variant model. The dynamic model parameters were determined via nonlinear modeling method, linear time invariant interpolator, and genetic algorithm. The angle estimator performs well and the linear parameter variant model efficiently estimates the micromanipulator’s behavior. The results presented here provide a workable foundation for surgical robot micromanipulator force estimation and control.

scholarly journals Effect of Time Delays in Characterizing Continuous Mixing of Aqueous Xanthan Gum Solutions

2021 ◽  
Author(s):  
Vishal Kumar Patel
Keyword(s):  
Dynamic Model ◽  
Discrete Time ◽  
Time Domain ◽  
Time Delays ◽  
Xanthan Gum ◽  
Dynamic Models ◽  
Hybrid Genetic Algorithm ◽  
Mixed Volume ◽  
Model Parameters ◽  
Continuous Mixing

Aqueous xanthan gum solutions are non-Newtonian fluids, pseudoplastic fluids possessing yield stress. Their continuous mixing is an extremely complicated phenomenon exhibiting non idealities such as channeling, recirculation and stagnation. To characterize the continuous mixing of xanthan gum solutions, three dynamic models were utilized: (1) a dynamic model with 2 time delays in discrete time domain, (2) a dynamic model with 2 time delays in continuous time domain, and (3) a simplified dynamic model with 1 time delay in discrete time domain. A hybrid genetic algorithm was employed to estimate the model parameters through the experimental input-output dynamic data. The extents of channeling and fully-mixed volume were used to compare the performances of these three models. The dynamic model parameters exerting strong influence on the model response were identified. It was observed that the models with 2 time delays gave a better match with the experimental results.

scholarly journals Dynamic Modeling of a Spatial Cable-Driven Continuum Robot Using Euler-Lagrange Method

2020 ◽  
Vol 10 (1) ◽  
pp. 60-74
Author(s):  
Ammar Amouri ◽  
Chawki Mahfoudi ◽  
Abdelouahab Zaatri
Keyword(s):  
Dynamic Model ◽  
Dynamic Modeling ◽  
Dynamic Models ◽  
Model Development ◽  
Mathematical Expression ◽  
Gravitational Energy ◽  
Lagrange Method ◽  
Continuum Robots ◽  
Highly Nonlinear ◽  
Continuum Robot

Continuum robots are kinematically redundant and their dynamic models are highly nonlinear. This study aims to overcome this difficulty by presenting a more practical dynamic model of a certain class of continuum robots called cable-driven continuum robot (CDCR). Firstly, the structural design of a CDCR with two rotational degrees of freedom (DOF) is introduced. Then, the kinematic models are derived according to the constant curvature assumption. Considering the complexity of the kinetic energy expression, it has been approximated by the well-known Taylor expansions.  This case corresponds to weak bending angles within the specified bending angle range of the robot. On the other hand, due to the low weight of the CDCR components, the gravitational energy effects can be neglected compared to those stemmed from the elastic energy. Thereafter, the corresponding dynamic model is established using Euler-Lagrange method. Static and dynamic models have been illustrated by examples. This analysis and dynamic model development have been compared with the existing scientific literature. The obtained results shown that the consistency and the efficiency of accuracy for real-time have been carried out. However, the dynamic modeling of CDCR with more than 2-DOF leads to a more complex mathematical expression, and cannot be simplified by adopting the similar assumptions and methodology used in the case of 2-DOF.

Optimal Web Guiding

2009 ◽  
Vol 132 (1) ◽  
Author(s):  
Aravind Seshadri ◽  
Prabhakar R. Pagilla
Keyword(s):  
Dynamic Model ◽  
Dynamic Models ◽  
Fiber Optic ◽  
Operating Conditions ◽  
Model Parameters ◽  
Linear Quadratic ◽  
Position Measurement ◽  
Series Of Experiments ◽  
Correct Application ◽  
Selection Of

This paper presents an optimal web guiding strategy based on the dynamic analysis of the lateral web behavior and a new fiber optic lateral web position measurement sensor. First, a lateral dynamic model of a moving web is revisited with an emphasis on correct application of appropriate boundary conditions. Then the dynamic models of two common intermediate guides (remotely pivoted guide and offset-pivot guide) are investigated. The effect of various model parameters on lateral web behavior is analyzed and discussions on proper selection of the parameters are given. Based on the model analysis, we discuss the design of a linear quadratic optimal controller that is capable of accommodating structured parametric uncertainties in the lateral dynamic model. The optimal guide control system is evaluated by a series of experiments on a web platform with different web materials under various operating conditions. Implementation of the controller with a new fiber optic lateral sensor for different scenarios is discussed. Results show good guiding performance in the presence of disturbances and with uncertainties in the model parameters.

scholarly journals Effect of Time Delays in Characterizing Continuous Mixing of Aqueous Xanthan Gum Solutions

2021 ◽  
Author(s):  
Vishal Kumar Patel
Keyword(s):  
Dynamic Model ◽  
Discrete Time ◽  
Time Domain ◽  
Time Delays ◽  
Xanthan Gum ◽  
Dynamic Models ◽  
Hybrid Genetic Algorithm ◽  
Mixed Volume ◽  
Model Parameters ◽  
Continuous Mixing

Aqueous xanthan gum solutions are non-Newtonian fluids, pseudoplastic fluids possessing yield stress. Their continuous mixing is an extremely complicated phenomenon exhibiting non idealities such as channeling, recirculation and stagnation. To characterize the continuous mixing of xanthan gum solutions, three dynamic models were utilized: (1) a dynamic model with 2 time delays in discrete time domain, (2) a dynamic model with 2 time delays in continuous time domain, and (3) a simplified dynamic model with 1 time delay in discrete time domain. A hybrid genetic algorithm was employed to estimate the model parameters through the experimental input-output dynamic data. The extents of channeling and fully-mixed volume were used to compare the performances of these three models. The dynamic model parameters exerting strong influence on the model response were identified. It was observed that the models with 2 time delays gave a better match with the experimental results.

Sign in / Sign up

Export Citation Format

Share Document