Evaluation of the Probabilistic Design Methodology and Computer Code for Composite Structures
Abstract
This report presents the results of an independent evaluation on the numerical accuracy and computational efficiency of a probabilistic design methodology for composite aircraft structures. The methodology was developed by Northrop-Grumman Commercial Aircraft Division (NGCAD) under the Federal Aviation Administration (FAA) funding through Interagency Agreement DTFA03-94-A-40021, while the associated PC-based computer code MONTE was developed through FAA Grant 96-G-0036 with the University of Texas at Arlington. The probability calculation of NGCAD's probabilistic methodology is based on the conditional expectation method (CEM) to determine the failure probability of a specified failure event. This methodology was first verified by traditional Monte Carlo simulation (MCS) method using the computer code NESSUS developed by the National Aeronautical and Space Administration (NASA). Since Monte Carlo simulation is not efficient for small probability calculation, a mixed probabilistic method (MPM) was developed in this study to verify the results from MONTE in the 10(exp -6) or less probability level. The mixed probabilistic method requires a decomposition of the overall failure function into several conditional failure functions. The probability of failure for each conditional failure function is first calculated using the CEM. The overall failure probability is then computed using the probability integration method (PIM). The mixed probabilistic method is implemented in the computer code NESSUS. Structural reliability analyses were conducted on the wing box of a Lear Fan aircraft using the computer codes MONTE and NESSUS with MCS and MPM. The results from all the codes were compared. The comparison indicates that NGCAD probabilistic composite design methodology uses fewer random simulations than that for traditional Monte Carlo simulation method for an accurate probability prediction. The computational time was reduced by about an order of magnitude.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2001
- Accession Number
- ADA392690
Entities
People
- Mimichael Shiao