Cyber-Engineering: Advances in Simulation and Visualization for Engineering Design

Transcription

1 November 11, 2007 ASME Winter Annual Meeting CDI Workshop

2 Presentation outline Setting the stage From data to knowledge Understanding complexity in natural, built, and social systems Building virtual organizations Closing thoughts

3 Cyber-Engineering Cyber-Engineering is intricately linked to two conclusions: the inclusion of ICT affects virtually all forms of scholarly activity, and the speed of scholarly research is increasing Research Opportunities in CyberEngineering and CyberInfrastructure Development, A Report on the 3 rd CyberInfrastructure Workshop at Drexel University, Philadelphia, 2004 Cyber implies high-tech or born of the information age Engineering with and through digital, virtual, and computing mediums

4

5 Presentation outline Setting the stage From data to knowledge Understanding complexity in natural, built, and social systems Building virtual organizations Closing thoughts

6 Multidimensional data visualization Some methods exist for medium-dimensional problems (Parallel coordinates, ATSV) n-d measurement methods for medium and large-dimensional problems

7 Intuitive dimension reduction A four-objective (attribute) problem representation Two objectives on each axis An eight-objective (attribute) problem representation. Four objectives on each axis

8 Preferences weights Four-objective problem: W 4 = 0.5 W 4 = 0.75 W 4 = 0.95

9 Preference ranges The visualization of the fiveobjective hyperspace Pareto frontier. Zoomed Region Data for points in zoomed region Region to zoom Color map of preference incorporation

10 Presentation outline Setting the stage From data to knowledge Understanding complexity in natural, built, and social systems Building virtual organizations Closing thoughts

11 Emergent social systems Emergency evacuation Vacate uses a modified particle-swarm optimization for evacuation simulation using the Fire Dynamics Simulator (NIST). Integrates faculty from social sciences, management (first responder), and mechanical engineering (optimization and fire propagation modeling). Multidisciplinary Optimization & Design Engineering Lab (MODEL)

12 Reconfigurable systems

13 Reconfigurable systems Reconfigurable (flexible) system design Capable of meeting multiple operating requirements after system deployment Systems designed to maintain a high level of performance through real time adaptations in their configuration and/or through robust parameter settings when operating conditions or requirements change in a predictable or unpredictable way

14 Reconfigurable in the literature

15 Dimensional flexibility Mass dimensional flexibility: the addition or removal of elements in the form of hardware, personnel, or software Length dimensional flexibility: geometric changes within the system after the system has been fielded Time dimensional flexibility: making decisions about the values of the design reducing the time required for redesign by utilizing an easily modified architecture

16 Case study Vehicle design Center of Gravity Cl R KR Straightaway Cl F Radius R1 140 h FRONT KF a l Radius R2 400 tf a a a' = = l a + b K' = C' = C K lf F C K + lf F + C K lr R Normalized center of gravity location Mass and Structures Normalized roll stiffness distribution Handling and Response Normalized aerodynamic downforce distribution Aerodynamics

17 Multiobjective capability Performance Space Design Space F 2 B Utopia Point X 2 Pareto Frontier B A A Primary research challenges: F 1 X 1 Minimize changes in design variables (optimization) Maintain best performance possible (cost/market modeling) Track or match resulting reconfiguration path (nonlinear controls, system identification, stability, mechanism design)

18 Cyber-enabled reconfigurable systems Opportunities for cyberinfrastructure-enabled reconfigurable systems. Design repositories Sensing technologies Data mining Networks/ Ambient intelligence Virtual simulation Control algorithms

19 Presentation outline Setting the stage From data to knowledge Understanding complexity in natural, built, and social systems Building virtual organizations Closing thoughts

20 The foundation: CIBER-U The Cyber-Infrastructure-Based Engineering Repositories for Undergraduates (CIBER-U): A collaborative online learning laboratory with cyberinfrastructure-based repositories and teaching materials to support educational initiatives and outreach rooted in engineering dissection.

21 CIBER-U: Current work

22 CIBER-HS: Outreach activities

23 CIBER-HS: Outreach activities Dissection activities Roller coaster repository components Virtual roller coaster designs Motion simulation experiences

24 Collaborators Virtual organization International Collaborators: UK Canada S. Korea Japan India Jordan Spain Germany Ireland Singapore

25 Product dissection VO

26 Presentation outline Setting the stage From data to knowledge Understanding complexity in natural, built, and social systems Building virtual organizations Closing thoughts

27 Closing thoughts - Transformations Mechanical engineering as the liberal arts of the 21 st century a foundation for cyber-engineering Data to knowledge: how to see and understand more information (operations research, data mining, mathematics) Understanding complexity: predictive reconfigurability (evolutionary computing, social science, multiobjective optimization, control theory, mechatronics) Virtual organization: product dissection (mechanical engineering, electrical engineering, computer science)

28 November 11, 2007 ASME Winter Annual Meeting CDI Workshop