Multibody Dynamics(2009)

所需积分/C币:9 2016-11-15 21:00:10 11.35MB PDF
收藏 收藏

Multibody Dynamics(2009)
Computational Methods in Applied Sciences Volume 12 Series editor E. Onate International Center for Numerical Methods in Engineering(CIMNE) Technical University of Catalunya (UPC) Edificio C-1, Campus Norte UPC Gran Capitan, S/n 08034 Barcelona, spain onate@@ cimne. upc. edu For other titles published in this series, go to Carlo L. bottasso Multibody Dynamics Computational Methods and applications Springer editor C.L. Bottasso Politecnico di milano Dipartimento di Ingegneria Aerospaziale Via La masa. 34 20156 Milano Italy carlo bottasso@polimiit ISBN978-1-4020-8828-5 e-ISBN978-1-4020-8829-2 Library of Congress Control Number: 2008933572 All Rights reserved C 2009 Springer Science Business Media B V. No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or b any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpos of being entered and executed on a computer system, for exclusive use by the purchaser of the work Printed on acid-free paper 987654321 springer. com Preface Multibody Dynamics is an area of Computational Mechanics which blends together various disciplines such as structural dynamics, multi-physics me- chanics, computational mathematics, control theory and computer science in order to deliver methods and tools for the virtual prototyping of complex mechanical systems. Multibody dynamics plays today a central role in the modeling, analysis, simulation and optimization of mechanical systems in a variety of fields and for a wide range of industrial applications The ECComas Thematic Conference on Multibody Dynamics was initi ated in Lisbon in 2003, and then continued in Madrid in 2005 with the goal of providing researchers in Multibody Dynamics with appropriate venues for exchanging ideas and results. The third edition of the Conference was held at the politecnico di Milano, Milano, Italy, from June 25 to June 28, 2007 The Conference saw the participation of over 250 researchers from 32 differ- ent countries, presenting 209 technical papers, and proved to be an excellent forum for discussion and technical exchange on the most recent advances in this rapidly growing field This book is a collection of revised and expanded versions of papers pre- sented at the Conference. Goal of this collection of works is to offer an up- to-date view on some of the most recent cutting edge research developments in Multibody dynamics. Contributions have been selected from all sessions of the Conference, and cover the areas of biomechanics(Ackermann and Schiehlen, Millard et al. ) contact dynamics(Tasora and Anitescu), control mechatronics and robotics(Bottasso), flexible multibody dynamics(Cugnon et al., Lunk and Simeon, Betsch and Sanger), formulations and numerical methods(Jay and Negrut), optimization( Collard et al. ) real-time simulation (Binami et al ) software development, validation, education(Pennestri and Valentini), and vehicle systems(Ambrosio et al. I hope you will find the reading of this collection enjoyable and stimulating as we anxiously wait for the 2009 edition of this excellent Conference Series Milano, May 2008 Carlo L. bottasso Contents Preface Physiological Methods to Solve the Force-Sharing Problem n biomechanics Marko ackermann and werner schiehlen Multi-Step Forward Dynamic Gait Simulation Matthew Millard. John McPhee and eric Kubica 25 A Fast NCP Solver for Large Rigid-Body Problems with Contacts. Friction. and joints alessandro tasora and mihai anitesca 45 Solution Procedures for Maneuvering Multibody Dynamics Problems for Vehicle Models of Varying Complexity Carlo L. bottasso 57 Synthesis and Optimization of Flexible mechanisms Frederic Cugnon, Alberto Cardona, Anna Selvi, Christian Paleczny, and martin Pucheta 81 The Reverse Method of Lines in Flexible Multibody dynamics Christoph Lunk and Bernd Simeon 95 a Nonlinear finite element framework for Flexible Multibody Dynamics: Rotationless Formulation and Energy-Momentum Conserving Discretization Peter Betsch and Nicolas sanger 119 a Second order extension of the generalized-a method for Constrained Systems in mechanics Laurent O.Jay and Dan Negrut 143 VII VIII Contents Kinematical Optimization of Closed-Loop Multibody Systems ean-Frangois Collard, Pierre duysin, and Paul Fisette 159 Dynamics Algorithms in Limited Parallel Computing A Comparison of Three Different Linear Order Multibod Environments Adarsh Binani, James H. Critchley, and Kurt S. Anderson 181 Linear Dual Algebra Algorithms and their Application to Kinematics Ettore Pennestri and pier paolo valentini 207 a Memory Based Communication in the Co-simulation of multibody and Finite Element Codes for Pantograph-Catenary Interaction Simulation Jorge Ambrosio, Joao Pombo, Frederico Rauter, and Manuel Pereira.. 231 Physiological methods to solve the Force-Sharing Problem in Biomechanics Marko Ackermann and Werner Schiehlen Department of Biomedical Engineering, Cleveland Clinic Foundation, 9500 Euclid Avenue/ND20, 44195 Cleveland, OH, USA E-mail: 2 Institute of Engineering and Computational Mechanics, University of Stuttgart Pfaffenwaldring 9, 70569 Stuttgart, Germany E-mail: schiehlen@itmuni-stuttgart. de Summary. The determination of individual muscle forces has many applications including the assessment of muscle coordination and internal loads on joints and bones, useful, for instance, for the design of endoprostheses. Because muscle forces cannot be directly measured without invasive techniques, they are often estimated from joint moments by means of optimization procedures that search for a unique solution among the infinite solutions for the muscle forces that generate the same joint moments. The conventional approach to solve this problem, the static opti mization, is computationally efficient but neglects the dynamics involved in muscle force generation and requires the use of an instantaneous cost function, leadin often to unrealistic estimations of muscle forces. An alternative is using dynamic optimization associated with a motion tracking, which is, however, computation- ally very costly. Other alternative approaches recently proposed in the literature are briefly reviewed and two new approaches are proposed to overcome the limitations of static optimization delivering more realistic estimations of muscle forces while being computationally less expensive than dynamic optimization 1 Introduction Inverse dynamics is used to compute the net joint moments required to gen- erate a measured motion. Although giving a clue about the intensity of the actuation required to accomplish the observed motion, net joint moments fail n delivering information on the forces applied by the individual muscles and other structures spanning the joints. Because the skeletal system is redun dantly actuated by muscles, i. e. there are many more muscles than actuated degrees of freedom, and many muscles are multi-articular, spanning more that one joint, the direct translation of net moments into muscle forces is not pos sible. Therefore, conclusions about muscle activity from net joint moments are not very reliable(Zajac et al. 33). Furthermore, the energy consumption C L Bottasso(ed ) Multibody Dynamics: Computational Methods and Applications, 1 C Springer Science+Business Media B V. 2009 M. Ackermann and W. Schiehlen involved, represented by the metabolic cost during human motion cannot be accurately assessed In order to solve the mathematically indeterminate problem and assess muscle forces, optimization approaches are employed. The classical static op timization approach is characterized by the search for muscle forces that mini- mize a cost function and fulfill constraints, given basically by bounded muscle forces and by the equations of motion or joint moments computed by inverse dynamics, respectively. The cost functions are mathematical expressions as sumed to model some physiological criteria optimized by the central nervous system during a particular activity In spite of being computationally efficient, the static optimization ap- proach assumes an instantaneous optimal distribution of muscle forces suffer- ing from two important limitations. Firstly, it neglects the muscle contraction and activation dynamics, what might lead to unphysiological estimations of muscle forces. Secondly, the cost functions must be an instantaneous measure of performance, what excludes the possibility of using time-integral criteria as for example total metabolic cost expended. The latter limitation is specially important for the analysis of human walking, since metabolic cost is accepted to play an important role during locomotion The muscle activation and contraction dynamics can be taken into account by using dynamic optimization associated with the tracking of the prescribed kinematics. This approach is based on the search for optimal controls, in this case the neural excitations, that drive a forward-dynamics model of the musculoskeletal system to track the prescribed motion. Due to the several numerical integrations of the differential equations necessary, a prohibitive computational effort is required to achieve a solution. This draw back prevents this approach from being widely used and stimulated recent efforts to reduce the computational burden. Some strategies based on dynamic optimization are presented in more details in Section 3.2 An approach to solve the distribution problem in biomechanics is proposed in Section 4, see also Ackermann and Schiehlen [3 It considers the muscle contraction and activation dynamics and permits the use of time-integral cost functions as the total metabolic cost. This approach is called extended inverse dynamics(EID) because it requires, in addition to the inversion of the skele tal system dynamics, the inversion of the muscle contraction and activation dynamics. Since no numerical integration of the differential equations is re- quired, the extended inverse dynamics is computationally less costly than the dynamic optimization. A second, simplified approach, called modified static optimization(MSO), to permit computation of muscle forces that fulfill the constraints given by the activation and contraction dynamics is also proposed and presented in Section 5. The latter approach maintains computational ef fort similar to the ones for static optimization, while considering the dynamics involved in the muscle force generation process

试读 127P Multibody Dynamics(2009)
立即下载 身份认证后 购VIP低至7折
关注 私信
Multibody Dynamics(2009) 9积分/C币 立即下载
Multibody Dynamics(2009)第1页
Multibody Dynamics(2009)第2页
Multibody Dynamics(2009)第3页
Multibody Dynamics(2009)第4页
Multibody Dynamics(2009)第5页
Multibody Dynamics(2009)第6页
Multibody Dynamics(2009)第7页
Multibody Dynamics(2009)第8页
Multibody Dynamics(2009)第9页
Multibody Dynamics(2009)第10页
Multibody Dynamics(2009)第11页
Multibody Dynamics(2009)第12页
Multibody Dynamics(2009)第13页
Multibody Dynamics(2009)第14页
Multibody Dynamics(2009)第15页
Multibody Dynamics(2009)第16页
Multibody Dynamics(2009)第17页
Multibody Dynamics(2009)第18页
Multibody Dynamics(2009)第19页
Multibody Dynamics(2009)第20页

试读结束, 可继续阅读

9积分/C币 立即下载