SENSITIVITY ANALYSIS FOR LARGE-DEFLECTION AND POSTBUCKLING RESPONSES ON DISTRIBUTED-MEMORY COMPUTERS

A computational strategy is presented for calculating sensitivity coef ficients for the non-linear large-deflection and postbuckling response s of laminated composite structures on distributed-memory parallel com puters. The strategy is applicable to any message-passing distributed computational environment. The key elements of the proposed strategy a re: (a) a multiple-parameter reduced basis technique; (b) a parallel s parse equation solver based on a nested dissection (or multilevel subs tructuring) node ordering scheme; and (c) a multilevel parallel proced ure for evaluating hierarchical sensitivity coefficients. The hierarch ical sensitivity coefficients measure the sensitivity of the composite structure response to variations in three sets of interrelated parame ters; namely, laminate, layer and micromechanical (fiber, matrix and i nterface/interphase) parameters. The effectiveness of the strategy is assessed by performing hierarchical sensitivity analysis for the large -deflection and postbuckling responses of stiffened composite panels w ith cutouts on three distributed-memory computers. The panels are subj ected to combined mechanical and thermal loads. The numerical studies presented demonstrate the advantages of the reduced basis technique fo r hierarchical sensitivity analysis on distributed-memory machines.