Cadalyst - September 2007 - (Page 14) techtrends finite-element program capable of simulating complex real-world problems.” LS-DYNA has been used to analyze heart valves, dental bridges, corrective orthopedic shoes, airbags, and train crashes. The Purdue team used this package to run the finite-element analysis (FEA) simulations of the attack on the WTC. “The FEA model [of the aircraft] includes structural elements, including ribs, stringers, keel beam, floor, and more,” the team wrote in its report. Ayhan Irfanoglu, assistant professor of civil engineering at Purdue and another member of the team, further explained, “Smoothed particle hydrodynamics (SPH) elements were used to model the jet fuel. We focused on the physics of the aircraft-impact problem, generalized as the impact of fluid traveling at high speed on solid. We tested our theoretical work using small-scale experiments.” Of the WTC model, the report notes, “All stories were modeled, including those underground. The simulation restricted to the upper 20 floors of the building, however, [was done] with increased detail meshing near the impact region so as to achieve high accuracy of the results.” “We didn’t have prior experience with meshing tools,” Hoffmann said, “so there was a fair amount of time spent just putting the models together [figure 2].” Figure 2. The FEA model for the North Tower included all stories, including those underground. However, the simulation restricted to the upper 20 floors — the impact region of the building — was done with increased detail meshing. Irfanoglu, and Mete Sozen) point out that the Riera approach is especially useful in the estimation of: mass distribution versus crushing force velocity and force distribution during impact impact damage without the detailed aircraft model For hard evidence, Hoffmann said he and his colleagues also “shot beverage cans into a stiff metal plate” to obtain data to compare with their FEA-based estimates. The Physics of Destruction What happens when an airplane collides with a solid wall? If you’re mathematically inclined, you’ll be able to dissect the energy distribution and the behavior of the impact forces using the Riera curve, an algorithmic summation of such incidents (figure 3). In one of the archived Purdue presentation documents on Riera curve, the authors (Christoph Hoffmann, Ayhan Still Unsettled But don’t think this simulation will answer all the questions that people have been raising about the collapse of the WTC. Even though they believe that for the aircraft and for the parts of the structure it modeled, their simulation is faithful to the physics of the problem, the team warns that “impact simulations indicate that the damage states of the WTC-I core structural elements are very sensitive to analysis parameters and as such, it is not possible to suggest the exact state of the core framing after the airWhat It Takes to Recreate craft impact.” a Single Second Hoffmann explained, “In other The video clip released by Purdue’s Rosen Center for Advanced Computing, a division words, we can’t say, ‘Well, this numof Information Technology at Purdue University, tracks one second of real time during ber of columns were cut; therefore, the moment of impact at WTC’s North Tower on September 11, 2001. The statistics the core was compromised.’ In fact, below reveal the complexity of the digital reenactment. in most of the simulation runs we did, The aircraft and the tower models: a total of 332,862 nodes after the impact, the core still had Smoothed particle hydrodynamics: 87,188 nodes plenty of capacity to carry the load Beams: 248,433 nodes [of the tower].” Shells: 93,733 nodes With scientific reservations, HoffSolid elements: 674 nodes mann was prepared only to suggest, Time required for FEA: 166 hours (on an IBM Regatta with 16 Power-5 processors) “The fire seemed to have been the Disk space: 20 GB for the simulation data files and the 400 saved states culprit in weakening the structure. 14 September 2007 | cadalyst | www.cadalyst.com http://www.cadalyst.com
Table of Contents Feed for the Digital Edition of Cadalyst - September 2007 Contents Editor's Window Techtrends: Resurrecting an Unforgettable Second Living Large UGS Solid Edge v20 Adobe Acrobat 3D v8 Adobe Photoshop CS3 Extended CAD Manager: Managing 2D and 3D in the Same Office MCAD Modeling: Where is MCAD Going? AEC in Focus: Automated Estimating and Scheduling Using BIM CAD Cartoon Issue Indexes CADfidential Cadalyst - September 2007 Cadalyst - September 2007 - (Page FC1) Cadalyst - September 2007 - (Page FC2) Cadalyst - September 2007 - (Page Cover1) Cadalyst - September 2007 - (Page Cover2) Cadalyst - September 2007 - (Page 3) Cadalyst - September 2007 - Contents (Page 4) Cadalyst - September 2007 - Contents (Page 5) Cadalyst - September 2007 - Contents (Page 6) Cadalyst - September 2007 - Contents (Page 7) Cadalyst - September 2007 - Editor's Window (Page 8) Cadalyst - September 2007 - Editor's Window (Page 9) Cadalyst - September 2007 - Editor's Window (Page 10) Cadalyst - September 2007 - Editor's Window (Page 11) Cadalyst - September 2007 - Techtrends: Resurrecting an Unforgettable Second (Page 12) Cadalyst - September 2007 - Techtrends: Resurrecting an Unforgettable Second (Page 13) Cadalyst - September 2007 - Techtrends: Resurrecting an Unforgettable Second (Page 14) Cadalyst - September 2007 - Techtrends: Resurrecting an Unforgettable Second (Page 15) Cadalyst - September 2007 - Living Large (Page 16) Cadalyst - September 2007 - Living Large (Page 17) Cadalyst - September 2007 - Living Large (Page 18) Cadalyst - September 2007 - Living Large (Page 19) Cadalyst - September 2007 - Living Large (Page 20) Cadalyst - September 2007 - Living Large (Page 21) Cadalyst - September 2007 - Living Large (Page 22) Cadalyst - September 2007 - Living Large (Page 23) Cadalyst - September 2007 - Living Large (Page 24) Cadalyst - September 2007 - Living Large (Page 25) Cadalyst - September 2007 - Living Large (Page 26) Cadalyst - September 2007 - Living Large (Page 27) Cadalyst - September 2007 - UGS Solid Edge v20 (Page 28) Cadalyst - September 2007 - UGS Solid Edge v20 (Page 29) Cadalyst - September 2007 - UGS Solid Edge v20 (Page 30) Cadalyst - September 2007 - UGS Solid Edge v20 (Page 31) Cadalyst - September 2007 - Adobe Acrobat 3D v8 (Page 32) Cadalyst - September 2007 - Adobe Acrobat 3D v8 (Page 33) Cadalyst - September 2007 - Adobe Acrobat 3D v8 (Page 34) Cadalyst - September 2007 - Adobe Photoshop CS3 Extended (Page 35) Cadalyst - September 2007 - Adobe Photoshop CS3 Extended (Page 36) Cadalyst - September 2007 - CAD Manager: Managing 2D and 3D in the Same Office (Page 37) Cadalyst - September 2007 - CAD Manager: Managing 2D and 3D in the Same Office (Page 38) Cadalyst - September 2007 - CAD Manager: Managing 2D and 3D in the Same Office (Page 39) Cadalyst - September 2007 - MCAD Modeling: Where is MCAD Going? (Page 40) Cadalyst - September 2007 - MCAD Modeling: Where is MCAD Going? (Page 41) Cadalyst - September 2007 - MCAD Modeling: Where is MCAD Going? (Page 42) Cadalyst - September 2007 - MCAD Modeling: Where is MCAD Going? (Page 43) Cadalyst - September 2007 - MCAD Modeling: Where is MCAD Going? (Page 44) Cadalyst - September 2007 - AEC in Focus: Automated Estimating and Scheduling Using BIM (Page 45) Cadalyst - September 2007 - AEC in Focus: Automated Estimating and Scheduling Using BIM (Page 46) Cadalyst - September 2007 - AEC in Focus: Automated Estimating and Scheduling Using BIM (Page 47) Cadalyst - September 2007 - AEC in Focus: Automated Estimating and Scheduling Using BIM (Page 48) Cadalyst - September 2007 - Issue Indexes (Page 49) Cadalyst - September 2007 - Issue Indexes (Page 50) Cadalyst - September 2007 - Issue Indexes (Page 51) Cadalyst - September 2007 - CADfidential (Page 52) Cadalyst - September 2007 - CADfidential (Page Cover3) Cadalyst - September 2007 - CADfidential (Page Cover4)
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.