Presentation on theme: "Carbon/Epoxy Laminate Compression After Impact Load Prediction from Ultrasonic C-Scan Data Eric v. K. Hill, Christopher D. Hess and Yi Zhao OBJECTIVES."— Presentation transcript:
Carbon/Epoxy Laminate Compression After Impact Load Prediction from Ultrasonic C-Scan Data Eric v. K. Hill, Christopher D. Hess and Yi Zhao OBJECTIVES Three sets of 3.5 x 6 inch 16-ply AS4/3501-5A carbon/epoxy coupons impacted from 0-20 ft-lb f with 5/8 inch diameter hemispherical tup to create barely visible impact damage (BVID) Back-propagation neural network (BPNN) prediction of compression after impact (CAI) load from transformed ultrasonic (UT) C-scan image Goal:±15%Goal: Worst case prediction error within ±15% APPROACH/TECHNICAL CHALLENGES AE data too noisy: Train BPNN using 50 data points representing column summation data from UT C-scan image and known CAI loads as input Test BPNN using column summation UT C-scan image to predict CAI loads on remaining coupons ACCOMPLISHMENTS/RESULTS worst case errors -12.12%, 16.62%, and -11.83% for the three setsUT image data alone used to predict ultimate compressive strengths with worst case errors of -12.12%, 16.62%, and -11.83% for the three sets predict accurately without known impact energyBPNN able to predict accurately without known impact energy – valid for real world applications such as impact damaged aircraft wings C/Ep Coupon in Boeing BS-7260 Compression After Impact Test Fixture with Three Acoustic Emission Transducers Attached Instron Dynatup 9250 Calibrated Impactor Delaminations in Coupon Due to Impact Damage
MATLAB Data Transformation Pixel color and location is represented by a matrix array of numbers (0-16) Numerical values represent hue color Image data summed and normalized in the column direction 50-100 data points surrounding the maximum used as inputs to BPNN UltraPAC II C-Scan Imaging System: Water Couplant Immersion 5 MHz Unfocused Transducer 16 Color Format 0-15 Color Format Digital Representation of 0-15 Color Format
Data Set SpecimenImpact Energy (ft-lb f ) Compressive Load (lb f ) Predicted Compressive Load (lb f ) % Error Training A20 2865.6 2865.600.00 A32.23 6531.9 6531.900.00 A521.43 3910.1 3910.100.00 A420.2 3042.4 3042.400.00 Testing A620.75 4174.8 4492.737.62 A10 4936.5 4338.07-12.12 BPNN Predictions for Batch A Coupons Optimized BPNN Settings Digital Ultrasonic C-Scan Image Data Predicted CAI Load NeuralWorks Professional II/PLUS ® Software Summary of BPNN Training and Test Results Worst Case Error
Your consent to our cookies if you continue to use this website.