Conference PaperPDF Available

Impact of defects during automated fibre placement on the compression behaviour of cured CFRP structures

Authors:

Abstract and Figures

Process induced defects during automated fibre placement (AFP) can have an impact on the mechanical behaviour of cured fibre composite structures. The influence of recurring defects such as gaps and overlaps have been extensively studied whereas the impact of tow-twists and fuzzballs is not well known. Hence, the findings of compression tests carried out on carbon fibre-reinforced polymer (CFRP) test specimens with intentionally placed tow-twists and fuzzballs are presented in this study. The defects are characterised based on geometric data acquired during layup. The fully cured defect specimens are further examined by computer tomography (CT), giving an unprecedented understanding of how the defects deform during curing and affect the neighbouring laminae. In addition, representative 3D meso-scale finite-element simulations of laminates containing the aforementioned defects are conducted, using the geometric data acquired by the CT-scans. The simulation output is then compared to the behaviour observed in the experimental tests. The results from the compression tests indicate an influence of the defects on the compression strength, compression stiffness and buckling behaviour of the laminate, depending on the characteristic defect size and defect position in the layup. A conclusion is drawn towards being able to accurately predict the mechanical impact of fuzzballs and tow-twists numerically, thus aiding in the decision-making of when to remove such defects during parts production in the AFP process.
Content may be subject to copyright.
SAMPE Europe Conference 2022 Hamburg - Germany
1
IMPACT OF AUTOMATED FIBRE PLACEMENT INDUCED
DEFECTS ON THE COMPRESSION BEHAVIOUR OF CFRP
STRUCTURES
!"#$%&'()$*%#%+a,(-%.&'/*&"(0%*1.'a,(2%/%$(03$'/a,(4&$'/%"(-561*#/b,(7&$5(
8*11%$1&""b,()*$&'(0&9*$*a,(!+%:&"#%$(0%$;*<a(
a(="'/*/>/%(3?(!*$5$&?/(@%'*<"(&"#(A*<6/;%*<6/(-/$>5/>$%',(8%56"*'56%(B"*C%$'*/D/(
E$&>"'56;%*<,(F//%".%59%$(@&11(GH,(-/&#%,(I%$1&"J(
b(="'/*/>/%(3?(2$3#>5/*3"(K"<*"%%$*"<(&"#(7&56*"%(833+',(A%*."*L(B"*C%$'*/D/(0&""3C%$,(
F//%".%59%$(@&11(GH,(-/&#%,(I%$1&"J((
ABSTRACT
2$35%''(*"#>5%#(#%?%5/'(#>$*"<(&>/31&/%#(?*.$%(M+&5%1%"/(N!)2O(6&C%(&"(*1M&5/(3"(/6%(
1%56&"*5&+(.%6&C*3>$(3?(5>$%#(?*.$%(531M3'*/%('/$>5/>$%'P(Q6*+%($%5>$$*"<(#%?%5/'('>56(
&'( <&M'( &"#( 3C%$+&M'( 6&C%( .%%"( %:/%"'*C%+J( '/>#*%#,( /6%( *1M&5/( 3?( /3;R/;*'/'( &"#(
?>LL.&++'(*'("3/(;%++(9"3;"P(0%"5%,(/6*'('/>#J(M$%'%"/'(/6%(?*"#*"<'(3?(531M$%''*3"(/%'/'(
5&$$*%#(3>/(3"(5&$.3"(?*.$%R$%*"?3$5%#(M3+J1%$(/%'/('M%5*1%"'(;*/6(*"/%"/*3"&++J(M+&5%#(
/3;R/;*'/'(&"#(?>LL.&++'P(86%(#%?%5/'(&$%(56&$&5/%$*'%#(.&'%#(3"(<%31%/$*5(#&/&(&5S>*$%#(
#>$*"<( +&J>MP( 86%( 5>$%#( 'M%5*1%"'( &$%( ?>$/6%$( %:&1*"%#( .J( 531M>/%$( /313<$&M6J,(
<*C*"<( &"( >"M$%5%#%"/%#( >"#%$'/&"#*"<( 3?( 63;( /6%( #%?%5/'( #%?3$1( #>$*"<( 5>$*"<( &"#(
&??%5/(/6%("%*<6.3>$*"<(+&1*"&%P(="(&##*/*3",(T@(1%'3R'5&+%(?*"*/%R%+%1%"/('*1>+&/*3"'(3?(
+&1*"&/%'(53"/&*"*"<(/6%(&?3$%1%"/*3"%#(#%?%5/'(&$%(53"#>5/%#(>'*"<(/6%(<%31%/$*5(#&/&(
?$31(/6%(/313<$&1'P(86%('*1>+&/*3"( 3>/M>/(*'(531M&$%#(/3(/6%( .%6&C*3>$( 3.'%$C%#(*"(
/6%(%:M%$*1%"/&+(/%'/'P(86%($%'>+/'(?$31( /6%( 531M$%''*3"(/%'/'(*"#*5&/%(&"(*"?+>%"5%(3?(
/6%(#%?%5/'(3"(/6%(531M$%''*3"('/$%"</6,(531M$%''*3"('/*??"%''(&"#(.>59+*"<(.%6&C*3>$(
3?( /6%( +&1*"&/%P( !( 53"5+>'*3"( *'( #$&;"( /3;&$#'( .%*"<( &.+%( /3( &55>$&/%+J( M$%#*5/( /6%(
1%56&"*5&+(*1M&5/(3?(?>LL.&++'(&"#( /3;R/;*'/'( ">1%$*5&++J,( /6>'( &*#*"<( *"( /6%(#%5*'*3"R
1&9*"<(3?(;6%/6%$(/3($%13C%('>56(#%?%5/'(#>$*"<(M&$/'(M$3#>5/*3"(*"(/6%(!)2(M$35%''P
(
1. INTRODUCTION
!>/31&/%#(?*.$%(M+&5%1%"/(*'(&(9%J(/%56"3+3<J(?3$('%$*&+(M$3#>5/*3"(3?(+&$<%(&"#(531M+%:(
5&$.3"(?*.$%($%*"?3$5%#(531M3'*/%(M&$/',(%'M%5*&++J(*"(/6%(&C*&/*3"(*"#>'/$J(UGVP(431M+%:(
*"/%$&5/*3"'(;*/6*"(/6%(+&J>M(M$35%''(63;%C%$(1&9%'(/6%(M$35%''(M$3"%(/3(%$$3$,(;6*56(
1*<6/( $%'>+/( *"( M$35%''( *"#>5%#( #%?%5/'( *"'*#%( /6%( 531M3'*/%( M&$/P( ->56( #%?%5/'( &$%(
/JM*5&++J(?*:%#( 1&">&++J( #>$*"<( /6%( +&J>M(M$35%'',( &'( &( 53$$%5/*3"( 3?( /6%( 5>$%#( M&$/( *'(
1>56(13$%(#*??*5>+/(&"#(53'/+JP(@%/%5/*3"(3?(#%?%5/'(#>$*"<(/6%(+&J>M(M$35%''(5&"($%'>+/(
*"(1&56*"%(#3;"/*1%(3?(>M(/3(WX(Y(UHVP(Z%'%&$56(*'(>"#%$;&J(/3(&>/31&/%(/6%(#%/%5/*3"(
3?(#%?%5/'(%*/6%$(/6$3><6(/3M3+3<J('5&""*"<(UTV(3$(.J(/6%$13<$&M6*5(1%&"'(UWVP(
Q6*+%(&>/31&/%#(*"'M%5/*3"(/%56"*S>%'(&*#(*"'M%5/3$'(/3(?*"#(&"#(5+&''*?J(#%?%5/'(#>$*"<(
/6%(+&J>M(M$35%'',(*/(*'('/*++(>M(/3(/6%(*"'M%5/3$(/3(#%5*#%(63;('%C%$%(&(#%/%5/%#(#%?%5/(*'(
&"#(/3(#%5*#%(;6%/6%$(*/("%%#'($%13C&+(3$("3/P(Z%'%&$56(*'('/*++("%%#%#(*"(3$#%$(/3(.%//%$(
S>&"/*?J(/6%(%??%5/(&(#%?%5/(6&'(3"(/6%(1%56&"*5&+(M$3M%$/*%'(3?(&(5>$%#(531M3'*/%(M&$/P(
!(C&$*%/J(3?(M$35%''(*"#>5%#(#%?%5/'(&$%(#*??%$%"/*&/%#(U[VP( 86%( %??%5/(3?(/3;(M+&5%1%"/(
%$$3$',( '>56( &'( <&M'( &"#( 3C%$+&M',( &$%( %:/%"'*C%+J( '/>#*%#( &'( /6%J( 3?/%"/*1%'( &$%(
SAMPE Europe Conference 2022 Hamburg - Germany
2
>"&C3*#&.+%(#>%(/3(M&$/(<%31%/$J(3$($%'/$*5/*3"'(*"(M$35%''(M&$&1%/%$'P(03;%C%$,(/6%$%(
&$%(#%?%5/'(;*/6(+%''%$(9"3;"(*1M&5/',('>56(&'(/;*'/%#(/3;'(3$(?>LL.&++'(N)*<>$%(GOP(86>',(
/6*'( '/>#J( &*1'( /3( %:M&"#( /6%( 9"3;+%#<%( 3?( /6%( 1%56&"*5&+( *"?+>%"5%(/3;R/;*'/'( &"#(
?>LL.&++'(%:%$/(3"(&(5>$%#(531M3'*/%(M&$/('>.\%5/%#(/3(531M$%''*C%(+3&#'P(
(
Figure 1: Tow-twist (left) and fuzzball (right)
2. EXPERIMENTATION
86$%%( /%'/( +&1*"&/%'( ;%$%( M$3#>5%#( ;*/6( &"( *"R63>'%( &>/31&/%#( ?*.$%( M+&5%1%"/(
1&56*"%(U]V,(?$31(;6*56(/%'/('M%5*1%"'(;%$%(5>/(3>/(&?/%$(5>$*"<P(86%(!)2(%"#(%??%5/3$(
*'(%S>*MM%#(;*/6(&"( *"?$&$%#( 5&1%$&( ?3$(*"R'*/>(/6%$13<$&M6*5(M$35%''( 13"*/3$*"<( UXVP(
!"( 3M/*3"&+(+&'%$( '5&""%$( 5&"(.%( ?*//%#( ?3$( &5S>*$*"<( &( 6%*<6/( 1&M( 3?( &( ?>++J( M+&5%#(
+&1*"&/%(+&J%$,(*"5+>#*"<(#%?%5/'P
(
!++( /%'/( +&1*"&/%'( ?%&/>$%#( &( +&J>M( 3?(G[( M+*%'P( 86%( ?*$'/( /%'/( +&1*"&/%( N2GO( ?%&/>$%'( &(
'J11%/$*5(5$3''(+&1*"&/%P(86%('%53"#(&"#(/6*$#(/%'/(+&1*"&/%'(N2H,(2TO(;%$%(#%'*<"%#(/3(
.%(S>&'*R*'3/$3M*5,( ;6*+%( /6%( /6*$#(/%'/(+&1*"&/%(6&'(13$%(#*$%5/*3"&+(M$3M%$/*%'(#>%(/3(
/6$%%(^_(+&J%$'(*"(*/'(5%"/$%P(86%(+&J>M(3?(/6%(/6$%%(/%'/(+&1*"&/%'(&$%(+*'/%#(*"(8&.+%(GP(
Table 1: Layup of the three test laminates used in experimentation
2G(
UN`^,(^O3,(`^,(^
aV'(
2H(
U^,(RW[,(^,(bW[,(^,(RW[,(bW[,(`^
a
a
a
a
V'(
2T(
U^,(RW[,(bW[,(RW[,(`^,(bW[,(^,(^
aV'(
(
2$35%''( *"#>5%#( #%?%5/'( ;%$%( M>$M3'%+J( *"'%$/%#( *"/3( /6%( +&1*"&/%( *"( #*??%$*"<( +&J%$'(
#>$*"<(/6%(+&J>M(M$35%''P(0%$%*",(+&J%$(G( $%?%$'( /3(/6%(.3//3113'/( M+JP( )>LL.&++'(;%$%(
+&*#(3"(/3M(3?(/6%('M%5*?*%#(+&J%$(&?/%$(;%*<6*"<P(83;R/;*'/'(;%$%(*"/%<$&/%#(.J(5&$%?>++J(
M>++*"<(/6%(53$$%'M3"#*"<(/3;(3>/(3?(/6%(!)2(6%&#,(/;*'/*"<(/6%(/3;,(&"#(/6%"(M>++*"<(/6%(
/3;(.&59(*"/3(/6%(6%&#(?3$(&>/31&/%#(M+&5%1%"/P(!+/63><6(1&">&++J(M+&5*"<(/6%(&??%5/%#(
/3;( ;*/6( /6%( /3;R/;*'/( 3"( /6%( 531M3'*/%( +&J>M( 1&J( .%( %&'*%$,( /6%( &?3$%1%"/*3"%#(
M$35%#>$%(.%//%$($%M$%'%"/'( /6%( 1&">?&5/>$*"<(M$35%'',(&'( /6%(53"/&5/(M$%''>$%( 3?( /6%(
!)2(531M&5/*3"($3++%$(*'("3/("%<+%5/%#P(4>$*"<(3?(/6%(/%'/(+&1*"&/%'(;&'(53"#>5/%#(*"(&"(
&>/35+&C%(;*/6('3?/(/33+*"<,(*P%P("3(6&$#(/33+*"<(;&'(M+&5%#(3"(/3M(3?(/6%(+&1*"&/%P
(
8%'/( 'M%5*1%"'( ;%$%( 5>/( ?$31( /6%( /%'/( +&1*"&/%'( ;*/6( &"( 3C%$&++( #*1%"'*3"( 3?(
G]^(11 c W^(11
P( 86%( 'M%5*1%"'( ;%$%( $%*"?3$5%#( ;*/6(
GP[(11
(/6*59( ?*.%$<+&''( 5&M(
'/$*M',(+%&C*"<(3M%"(&(/%'/(&$%&(3?(
W^(11 c W^(11
P(86%(#%?%5/'(*"'*#%(/6%('M%5*1%"'(;%$%(
C&$*%#(*"('*L%(&"#(+&J%$(3?(*"'%$/*3"(.&'%#(3"(&"(#%'*<"(3?(%:M%$*1%"/'P()3$(%&56(#%'*<"(
M3*"/,(/6$%%('M%5*1%"'(;%$%(5>/(3>/P((
SAMPE Europe Conference 2022 Hamburg - Germany
3
2.1 Naming convention
0%$%*"&?/%$,(/6%(#%'*<"(M3*"/'('6&++(.%($%?%$%"5%#(?3++3;*"<(/6%("&1*"<(53"C%"/*3"(
22d
@dAAd-
,(;6%$%(
22
($%?%$'( /3( /6%(/%'/( +&1*"&/%( N2G,(2H,( 2TO,(
@
(*"#*5&/%'(/6%( #%?%5/( /JM%(
N83;R/;*'/(3$()>LL.&++O,(
AA
(*'(/6%(M+J,(*"(;6*56(/6%(#%?%5/(;&'(*"'%$/%#,(&"#(
-
(*"#*5&/%'(
/6%( #%?%5/( '*L%( N-1&++( 3$( A&$<%OP( )>LL.&++'( &$%( 53"'*#%$%#( '1&++( ;6%"( /6%*$( C3+>1%( *'(
'1&++%$( /6&"(
HG^(11!
P( 83;R/;*'/'( &$%( 53"'*#%$%#( '1&++( *?( /6%*$( /3/&+(+%"</6( #3%'( "3/(
%:5%%#(
T^(11
P()3$(.%//%$(5+&$*/J,(/6%(3$*%"/&/*3"(3?(/3;R/;*'/'(*'(&+'3(<*C%"P(
2.2 Data processing
86%(1%56&"*5&+(M$3M%$/*%'(3?(/6%(0%:2+J(e[[HR!-W(>"*#*$%5/*3"&+(M$%M$%<(1&/%$*&+(;%$%(
#%/%$1*"%#( *"( M$%+*1*"&$J( /%'/'( ?3++3;*"<( '/&"#&$#'( @=f( Kf( =-F( [HXR[gH^^`( ?3$( /6%(
/%"'*+%( M$3M%$/*%',( @=f( Kf( =-F( GWGH]gH^^^( ?3$( /6%( 531M$%''*C%( M$3M%$/*%'( &"#( @=f(
][W]]gG``](?3$(/6%('6%&$(M$3M%$/*%'P(86%(&C%$&<%(5>$%#(M+J(/6*59"%''(;&'(?3>"#(/3(.%(
h"#$ iGeHPH j XPX(k1
( ;*/6( &"( &C%$&<%( ?*.$%( C3+>1%( ?$&5/*3"( 3?(
l
%m]H(Y
P( 86%( M+J(
M$3M%$/*%'( ;%$%( &5S>*$%#( .J( &"&+J'*"<( 1*5$3<$&M6'( 3?( 'M%5*1%"( "3/( 53"/&*"*"<( &"J(
#%?%5/'P
(
4>$%#(/%'/('M%5*1%"'(;*/6(*"6%$%"/(?>LL.&++'(;%$%('5&""%#(.J(531M>/%$(/313<$&M6J(&/(
/6%(="'/*/>/%(?3$(7&/%$*&+(-5*%"5%(&/(/6%(A%*."*L(B"*C%$'*/J(0&""3C%$(.%?3$%(531M$%''*3"(
/%'/*"<P(86%(/313<$&1'(?%&/>$%(&($%'3+>/*3"(3?(
GWPG[ k1 C3:%+
nP(@&/&(M3'/RM$35%''*"<(;&'(
#3"%(>'*"<(/6%('3?/;&$%(FZ-(@$&<3"?+J(UeVP(!(6%*<6/(1&M(3?(%&56(?>LL.&++(;&'(%:/$&5/%#(
C*&('%<1%"/&/*3"( N)*<>$%(HOP(@%'5$*M/*C%(M&$&1%/%$'('>56(&'(C3+>1%,(+%"</6,(;*#/6( &"#(
/6*59"%''(3?(/6%(?>LL.&++'(53>+#(/6%"(.%(#%$*C%#(?$31(/6%'%(U`VP(
(
Figure 2: 3D rendering of a segmented fuzzball. Pores inside the fuzzball are marked in red.
K:M%$*1%"/&/*3"(;&'( 5&$$*%#(3>/( &/(I$&''%( o>$( 431M3'*/%( 8%'/*"<( *"( E%$+*"( >/*+*L*"<( &(
6J#$&>+*5( 531M3'*/%( 531M$%''*3"( ?*:/>$%( ;*/6( +*"%&$( <>*#&"5%P( 86%( %+3"<&/*3"( 3?( &(
'M%5*1%"(;&'(1%&'>$%#(>'*"<(/6%(3M/*5&+(?*%+#('/$&*"(1%&'>$%1%"/('J'/%1(IF7(!$&1*'(
UG^VP(86*'(&++3;%#(/3(&55>$&/%+J(1%&'>$%(/6%('/$&*"(?*%+#(3?(/6%('M%5*1%"p'(>MM%$('>$?&5%(
*"(T@P(86%(+3&#(*'(*"/$3#>5%#('3+%+J(/6$3><6('6%&$R+3&#*"<(&+3"<'*#%(/6%(5&M('/$*M'P(86%(
<+3.&+('>$?&5%('/$&*"(
q&'(%)*
(*'(&5S>*$%#(.J(&C%$&<*"<(/6%(1%&'>$%#('/$&*"(?*%+#P(=/('63>+#(
.%( "3/%#( /6&/( /6%( 1%&"( '>$?&5%( '/$&*"(
q&'(%)*
( #%C*&/%'( ?$31( /6%( /6$3><6R/6*59"%''(
&C%$&<%#('/$&*"(
q*
(;*/6*"(/6%(531M3'*/%(M&$/(#>$*"<(.%"#*"<,(&'(3"+J(/6%(>MM%$('>$?&5%(
'/$&*"(3?(/6%('M%5*1%"(*'($%53$#%#P(
SAMPE Europe Conference 2022 Hamburg - Germany
4
3. RESULTS
86%( '/*??"%''( 3?( /6%( /%'/( 'M%5*1%"'( ;&'( 5&+5>+&/%#( &'( /6%( <$&#*%"/( 3?( /6%( '/$%''( &/( /6%(
%+3"<&/*3"'(
q+,+-. i ^P^[Y
(&"#(
q+,/-. i ^PH[Y
,(&'(#%'5$*.%#(*"(@=f(Kf(=-F(GWGH]gH^^^P(
86%( KR13#>+%'( &'( %:M%5/%#( ?$31( 5+&''*5&+( +&1*"&/*3"( /6%3$J(
r012
( &"#( /6%( 1%&"( KR
13#>+%'(&5S>*$%#(?$31(/6%(#%?%5/R?$%%($%?%$%"5%(/%'/'(
r34%
(&$%('63;"(*"(8&.+%(H(?3$(%&56(
/%'/( +&1*"&/%P( !'( *'( &MM&$%"/,( 5+&''*5&+( +&1*"&/*3"( /6%3$J( /%"#'( /3( 3C%$%'/*1&/%( /6%(
531M$%''*3"('/*??"%''(3?(&++(/6$%%(/%'/(+&1*"&/%'(;*/6(>M(/3(
TPHW(Y
P(
(
Table 2: E-module from classical lamination theory and mean E-module from tests of the three test laminates
s567(
s89:
[ePX]G(I2&(
[]P`G] j HP[^e(I2&(
[XP[XW(I2&(
[[PeTW j GPXGe(I2&(
[HPG]H(I2&(
[GP[H] j GPeGe(I2&(
(
!+13'/(&++('M%5*1%"',(%'M%5*&++J(?$31(/%'/(+&1*"&/%(2G,('63;(&(/%"#%"5J(/3(.>59+*"<(;%++(
.%?3$%($%&56*"<(/6%*$(1&:*1>1('/$%"</6(
t;<*
P(!('M%5*1%"(*'(53"'*#%$%#(/3(.>59+%(*?(/6%(
'/$%''R'/$&*"R5>$C%(5$3''%'( /6%( %R13#>+%(3??'%/( .J( R
^PG(Y
P(f3/%( /6&/( >M(/3( /6%( M3*"/(3?(
?*.$%(?&*+>$%,(/6%('M%5*1%"(.%6&C%'(%+&'/*5,(*P%P(/6%$%(*'("3(M+&'/*5(#%?3$1&/*3"(#%'M*/%(/6%(
&MM&$%"/(?+&//%"*"<(3?(/6%('/$%''R'/$&*"R5>$C%(N)*<>$%(TO,(;6*56($%'>+/'(?$31(.>59+*"<(&"#(
/6%(&?3$%1%"/*3"%#(3"%R'*#%#($%53$#*"<(3?(/6%('/$&*"(?*%+#P(
(
Figure 3: Stress-strain diagram for P1-REF-4. Local strain fields are shown at four distinct points during testing.
@*'$%<&$#*"<(.>59+*"<,(&+13'/(&++(/%'/('M%5*1%"'(?&*+%#(;*/6(&55%M/%#(13#%'(3?(?&*+>$%(&'(
#%'5$*.%#(*"(@=f(Kf(=-F(GWGH]gH^^^,(*P%P(/6%(M3*"/(3?(?&*+>$%(*'(+35&/%#(*"(/6%(5%"/$%(3?(/6%(
/%'/( &$%&( &"#( /6%( 'M%5*1%"'( /%"#( /3( 'M+*5%( *"( .3/6( #*$%5/*3"'( N)*<>$%( WOP( =11*"%"/(
#%+&1*"&/*3"'( ;%$%( 5+%&$+J( &>#*.+%( 131%"/'( .%?3$%( /3/&+( ?&*+>$%( 3?( /6%( 531M3'*/%(
+&1*"&/%P(
SAMPE Europe Conference 2022 Hamburg - Germany
5
(
Figure 4: Broken test specimen after compression testing.
Figure 5: Simulation boundary conditions
3.1 Simulation
="(5&'%(3?(9*"9*"<,(/6%(3.'%$C%#(1&:*1>1('/$%"</6(
t;<*
(?3$(&(/%'/('M%5*1%"(#*#("3/($%&56(
/6%( /6%3$%/*5&++J( %:M%5/%#(C&+>%P( -*1>+&/*3"( $>"'( ;*/6( *"5$%&'*"<( 531M$%''*3"( ?3$5%'(
;%$%(5&$$*%#(3>/(?3$(/6%($%?%$%"5%(/%'/',(*"(;6*56(/6%(+&1*"&/%(;&'(%C&+>&/%#(>'*"<(/6%(
?&*+>$%(13#%+(3?(2>59(UGGVP(86%(T@(?*"*/%(%+%1%"/(13#%+(;&'(M$%RM$35%''%#(;*/6(&(MJ/63"(
'5$*M/( /3( *"53$M3$&/%(/6%( #%?%5/( 56&$&5/%$*'/*5',( ;6*56( *'( #%/&*+%#( *"( U`VP( 86%(
$%M$%'%"/&/*C%( 13#%+( '*L%( ;&'( '%/( /3( 1&/56( /6%( 'M%5*1%"'( ?$%%( /%'/( &$%&,( *P%P(
W^(11 c W^(11
P(86%(M+J(/6*59"%''(*"(/6%('*1>+&/*3"(;&'(&$.*/$&$*+J('%/(/3(
h"#$ iGX[(k1
P(
86%(.3//31(+3&#*"<('>$?&5%(;&'(5+&1M%#(*"('M&5%(Nu%"5&'/$%vOP(86%(/3M(+3&#*"<('>$?&5%(
;&'($%'/$*5/%#(/3(3"+J(13C%(*"(:R#*$%5/*3"(/3(&553>"/(?3$(/6%(+*"%&$(<>*#&"5%(3?(/6%(/%'/(
'%/>M(&"#(&MM3*"/%#(&(531M$%''*C%(+3&#(N)*<>$%([OP(86%(+3&#(;&'(*"5$%&'%#(*"(*"5$%1%"/'(
3?(
^P[(9f
(>"/*+(?*.%$(?&*+>$%(;&'(#%/%5/%#(;*/6*"(/6%(+&1*"&/%P(
Table 3: Maximum strength values as measured in experiments and as expected from simulations for the three test laminates
(
K:M%$*1%"/&/*3"(
-*1>+&/*3"(
(
w=>?)89:
(
w@
(
wAA
(
wBAA
(
2G(
TX^P[ j GHPG(72&(
T[[PW j HTPG(72&(
]GWPT(72&(
x(
2H(
WXXPH j GTPT(72&(
WHGPH j HePG(72&(
[^`P[(72&(
T`[PH(72&(
2T(
WGGPT j GTPH(72&(
T][P` j GeP[(72&(
W]GP`(72&(
T[XPG(72&(
(
8&.+%(T(531M&$%'(/6%(%:M%$*1%"/&+(1&:*1>1('/$%"</6'(
t;<*)34%
(?$31(/6%($%?%$%"5%(/%'/'(
/3(/6%('/$%''%'(
tCC
,(&/(;6*56( ?*.$%(?&*+>$%(?*$'/( 355>$$%#( *"(/6%('*1>+&/*3"',( &"#(
tDCC
,(&/(
;6*56( *"/%$R?*.$%( ?&*+>$%,( *P%P( ?&*+>$%( 3?( /6%( 1&/$*:,( ?*$'/( 355>$$%#( *"( /6%( '*1>+&/*3"'P(
)>$/6%$13$%,(/6%('/$%''(
tE
(&/( ;6*56(/6%('M%5*1%"'('/&$/%#(/3(.>59+%(*'(&+'3(<*C%"P(f3/%(
/6&/(%'M%5*&++J(/6%( /%'/( +&1*"&/%(2G(*'( M$3"%( /3(9*"9*"<(;*/6(
t;<*)34% x tEiG[PG(72&
P(
E>59+*"<(;&'("3/( 3.'%$C%#(*"( /6%('*1>+&/*3"',( &'(/6%( '*1>+&/*3"'( ;%$%( '3+C%#( +*"%&$+J(
;*/6(&"(%:M+*5*/('3+C%$P((
SAMPE Europe Conference 2022 Hamburg - Germany
6
3.2 Test laminate P1
86%( /%'/( $%'>+/'( 3?( /6%( 'M%5*1%"'( ?$31( /%'/( +&1*"&/%( 2G( &$%( '63;"( *"( )*<>$%( ]P( 86%(
'M%5*1%"'( 3?( /%'/( +&1*"&/%( 2G( '63;%#( /6%( <$%&/%'/( /%"#%"5J( /3( 9*"9*"<P( -M%5*1%"'(
?%&/>$*"<(?>LL.&++'(<%"%$&++J('63;(&(#%5$%&'%(*"(/3+%$&.+%(531M$%''*3"('/$%''(
t;<*
P(-1&++(
?>LL.&++'(6%$%.J(/%"#(/3(6&C%(&(<$%&/%$(*1M&5/(3"(/6%(531M$%''*3"('/$%"</6P(-M%5*1%"'(
;*/6(?>LL.&++'(?>$/6%$('63;(&(53$$%+&/*3"(.%/;%%"(/6%(M3'*/*3"(3?(/6%(?>LL.&++(;*/6*"(/6%(
+&J>M( &"#( /6%( '/*??"%''(
r
P( Q6%"( ?>LL.&++'(&$%( *"/$3#>5%#( *"(+3;%$( M+*%',( /6%J(/%"#( /3(
#%5$%&'%('/*??"%'',(;6%$%&'(?>LL.&++'(M+&5%#(6*<6%$(>M(*"(/6%('/&59('63;(&"(*"5$%&'%(*"(
'/*??"%''P(
(
Figure 6: Deviation of
𝜎!"#
and
𝐸
relative to P1-REF for test laminate P1.
-M%5*1%"'(;*/6( /;*'/%#(/3;'(<%"%$&++J('63;(&"(*"5$%&'%(*"(.3/6(531M$%''*3"('/*??"%''(
&"#( '/$%"</6( ;6%"( 3$*%"/&/%#( M&$&++%+( /3( /6%( #*$%5/*3"( 3?( 531M$%''*3"P( 2%$M%"#*5>+&$(
+3&#*"<('63;'(3"+J(C%$J('1&++(#%5$%&'%(*"('/$%"</6(&"#('/*??"%''P((
83;R/;*'/'(;%$%("3/('*1>+&/%#P(I%"%$&++J,(/6%(*"?+>%"5%(#%?%5/'(6&C%(3"(/6%(1%56&"*5&+(
M$3M%$/*%'(3?(/6%(531M3'*/%(+&1*"&/%(*'(531M&$&.+J( '1&++( &"#(13'/+J($%'*#%(*"'*#%(/6%(
C&$*&/*3"(3?(/6%($%?%$%"5%('M%5*1%"'P((
3.3 Test laminate P2
)>LL.&++'(&+'3('63;(&"(*1M&5/(3"(/6%(1%56&"*5&+(.%6&C*3>$(3?(/%'/(+&1*"&/%(2H,(?%&/>$*"<(
&"(*"5$%&'%(*"('/$%"</6(?3$(/6%(+&$<%(?>LL.&++(*"/$3#>5%#(*"(+&J%$(X(N)*<>$%(XOP(86*'(1&J(
.%( *"/%$M$%/%#( '>56( &'( /6&/( ?>LL.&++'( *"( 1*##+%( +&J%$'( '/$%"</6%"( /6%( 531M&5/*3"(
.%6&C*3>$,(*P%P( /6%( ?>LL.&++(1&/%$*&+(1&J('>MM$%''(9*"9*"<(.J(/6*59%"*"<(/6%(531M3'*/%P(
SAMPE Europe Conference 2022 Hamburg - Germany
7
86*'( ?*"#*"<( 5&"( &+'3( .%( ?3>"#( *"( /6%( 3>/+*%$( 2GR)RAXRA( *"( /6%( $%'>+/'( 3?(/6%( ?*$'/( /%'/(
+&1*"&/%( 2GP( 86%( '1&++%$( ?>LL.&++'( *"( /6%( /%'/( +&1*"&/%( H( 63;%C%$( 3"+J( '63;( &( $&/6%$(
'1&++(#%C*&/*3"(*"('/$%"</6(531M&$%#(/3(/6%($%?%$%"5%('M%5*1%"'P(86%('/*??"%''(&<&*"(*'(
*"?+>%"5%#(/6%(13'/(/6$3><6(/6%(+&J%$*"<(M3'*/*3"(;*/6*"(/6%(531M3'*/%P(
Z%<&$#*"<(/3;R/;*'/',(/%'/( +&1*"&/%(2H('63;'( +*//+%(%??%5/(3"( /6%(531M$%''*3"('/*??"%''(
%:5%M/( ?3$( /6%( RW[_( 3$*%"/%#( /3;R/;*'/( *"( +&J%$( HP( I%"%$&++J,( /3;R/;*'/'( *"( &(
j
W[_(
3$*%"/&/*3"('%%1(/3('/$%"</6%"(/6%(+&1*"&/%,(;6*56(*'(*"(53"/$&#*5/*3"(/3(/6%(?*"#*"<'(*"(
/%'/(+&1*"&/%(2GP(
(
Figure 7: Deviation of
𝜎!"#
and
𝐸
relative to P2-REF for test laminate P2.
3.4 Test laminate P3
86%( %??%5/'( 3?( ?>LL.&++'( 3"( /6%( /%'/( +&1*"&/%( 2T( &$%( '31%;6&/( >"5+%&$( &"#( *"( M&$/(
53"/$&#*5/(?*"#*"<'( 3?( 2G( &"#( 2HP( 86%( +&$<%$( ?>LL.&++'( *"#*5&/%( &( #%M%"#%"5J( 3?( /6%(
531M$%''*3"('/$%"</6(/3(/6%(M+J(+&J%$,(*"(;6*56(/6%(?>LL.&++(;&'(M+&5%#P(86%('M%5*1%"'(
;*/6(&('1&++(?>LL.&++(*"(+&J%$(X(63;%C%$('63;(&"(>"%:M%5/%#(.%6&C*3>$,(&'(/6%J('63;(&(
<$%&/(*"5$%&'%( .3/6(*"( '/$%"</6(&"#( '/*??"%''P(863><6( /6%(#%?%5/( *'(M+&5%#( .%/;%%"(/6%(
/6$%%(&#\&5%"/(^_R+&J%$'( 3?( /6%(531M3'*/%,('3( &$%(/6%(?>LL.&++'( *"( 'M%5*1%"'(2TR)RAeRA,(
;6*56(63;%C%$(3"+J('63;(&('+*<6/(*"5$%&'%(*"('/$%"</6(&"#('/*??"%''P((
86%(/3;R/;*'/'( 3?(/%'/( +&1*"&/%( 2T(53"/$&#*5/( ?*"#*"<'(3?( /6%(3/6%$( /;3( /%'/( +&1*"&/%'P(
I%"%$&++J,(&++('M%5*1%"'(;*/6(*"6%$%"/(#%?%5/'(/%"#(/3(.3/6('/*??%"(&"#('/$%"</6%"(/6%(/%'/(
+&1*"&/%P(86%(*"5$%&'%(*"('/*??"%''(1&J(.%(#%M%"#%"/(3"(/6%(+&J%$(*"(;6*56(/6%(/3;R/;*'/(
SAMPE Europe Conference 2022 Hamburg - Germany
8
355>$$%#P(86*'(5&"(&+'3(.%(3.'%$C%#(*"(/6%($%'>+/'(3?(/%'/(+&1*"&/%(2GP(03;%C%$("3(5+%&$(
53$$%+&/*3"(;&'(?3>"#(?3$(/6%('/$%"</6(.%6&C*3>$(3?(/6%(/3;R/;*'/'(*"(/%'/(+&1*"&/%(2TP((
(
Figure 8: Deviation of
𝜎!"#
and
𝐸
relative to P3-REF for test laminate P3.
4. CONCLUSIONS
86%(&5S>*$%#(#&/&.&'%(3?(/6%(#%?%5/'(*'(>"*S>%,(&'(/6%(#%?%5/(<%31%/$J(3?(%&56(#%?%5/(;&'(
5&M/>$%#(.3/6(.%?3$%(&"#( &?/%$( &>/35+&C%(5>$*"<P(86*'(&++3;'( ?3$(&(#%/&*+%#(*"'*<6/( *"/3(
63;(#%?%5/'("3/(3"+J( $%&5/(/3(/6%(&>/35+&C%(M$%''>$%( 5J5+%,( .>/(&+'3(&??%5/('>$$3>"#*"<(
+&J%$'(;*/6*"(/6%(531M3'*/%(+&J>MP()>$/6%$13$%,(/6%$13<$&M6*5(*1&<%'($%53$#%#(#>$*"<(
+&J>M( &++3;( ?3$( &"( %:/%"'*C%( #&/&.&'%( /3( .%( .>*+#( >M( &"#( >'%#( ?3$( /$&*"*"<( 1&56*"%(
+%&$"*"<(.&'%#( 5+&''*?*%$'P( 86%($%'>+/'( ?$31(%:M%$*1%"/&/*3"( '>MM+%1%"/(/6%( #&/&.&'%,(
&++3;*"<(/3(M$%#*5/(/6%(1%56&"*5&+(*1M&5/(3?(#%?%5/'(3"+*"%(#>$*"<(/6%(+&J>M(M$35%''P(
86%(?*"*/%(%+%1%"/('*1>+&/*3"'(63;%C%$(#*#("3/(M$%#*5/(/6%(.%6&C*3>$(3?(/6%('M%5*1%"'(
#>$*"<(%:M%$*1%"/&/*3"P(86*'(*'(1&*"+J(#>%(/3(/6%(13#%+("3/(/&9*"<(.>59+*"<(*"/3(&553>"/P(
!+'3,( *11*"%"/( #%+&1*"&/*3"',( ;6*56( ;%$%( 5+%&$+J( &>#*.+%( #>$*"<( /%'/*"<,( ;%$%("3/(
&MM&$%"/(*"(/6%('*1>+&/*3"'P(86*'(.%6&C*3>$(1*<6/(.%(*"53$M3$&/%#(*"(?>/>$%('*1>+&/*3"'(
.J( *1M+%1%"/*"<( 536%'*C%( %+%1%"/'( *"( .%/;%%"( /6%( %+%1%"/'( 3?( &#\&5%"/( M+*%'P(
!##*/*3"&++J,(?>/>$%('*1>+&/*3"($>"'('63>+#(>'%(&"(*1M+*5*/('3+C%$(/3(&553>"/(?3$(.>59+*"<P((
KC&+>&/*3"( 3?( '*1>+&/*3"'( ;*/6( *"6%$%"/( ?>LL.&++'( *1M+*5&/%( &( $%#>5/*3"( 3?( /6%( 3C%$&++(
531M3'*/%('/$%"</6(#>%(/3(#%?%5/'P(86*'(*'(1&*"+J(#>%(/3(/6%( *"#>5%#( 3>/R3?RM+&"%( ?*.$%(
;&C*"%''( 5&>'%#( .J( /6%( ?>LL.&++',( ;6*56( *'( %:*'/%"/( *"( &++( +&J%$'( &.3C%( /6%( #%?%5/P(
f&/>$&++J,(*?(/6%(#%?%5/(*'(M+&5%#(*"(+3;%$(M+*%',(/6%(%??%5/(3?(/6%(#%?%5/(*'(/6>'(M$%#*5/%#(/3(
SAMPE Europe Conference 2022 Hamburg - Germany
9
.%( 13$%( '%C%$%P( 03;%C%$,( %:M%$*1%"/&/*3"( '63;%#( /6&/( #%?%5/'( 1*<6/( %C%"( 6&C%( &(
M3'*/*C%( %??%5/( 3"( /6%( 531M$%''*3"( '/$%"</6( 3?( /6%( 531M3'*/%,( &'( /6%J( 1*<6/( 6*"#%$(
.>59+*"<($%'M3"'%(3?(/6%(531M3'*/%(M+&/%P(
86%( &>/63$'( ;3>+#( +*9%( /3( 6*<6+*<6/( /6&/( %:M%$*1%"/&/*3"( *"( /6*'( '/>#J( 53"/$&#*5/'(
?*"#*"<'(3?(4$3?/(%/(&+P(UGHV(&"#(3?(0&$*9(%/(&+P(U[V,(;6*56(.3/6(?3>"#(/6&/(/3;R/;*'/'(6&C%(
&("%<&/*C%(*1M&5/(3"(/6%(531M$%''*3"('/$%"</6(3?(531M3'*/%(M+&/%'P(86*'('/>#J('><<%'/'(
/6&/( *?( &( 531M3'*/%( M+&/%( *'( M$3"%( /3( .>59+*"<,(/6*59"%''R&+/%$*"<( #%?%5/'( '>56( &'( /3;R
/;*'/'(1*<6/(6*"#%$(.>59+*"<($%'M3"'%(&"#(/6>'(&5/>&++J(5&"(6&C%(&(M3'*/*C%,(&+.%*/(C%$J(
'1&++,(%??%5/(3"(/6%(531M$%''*C%('/*??"%''(&"#('/$%"</6(3?(&(531M3'*/%('/$>5/>$%P
(
REFERENCES
[1] Brasington A, Sacco C, Halbritter J, Wehbe R, Harik R. Automated fiber placement: A
review of history, current technologies, and future paths forward. Composites Part C:
Open Access 2021;6:100182. https://doi.org/10.1016/j.jcomc.2021.100182.
[2] Schlegel K. A literature review of quality control for automated lay-up processes.
Zeitschrift Kunststofftechnik 2019;1:392–436. https://doi.org/10.3139/O999.01062019.
[3] Sacco C, Radwan AB, Harik R, Van Tooren M. Automated Fiber Placement Defects:
Automated Inspection and Characterization, Long Beach, CA: 2018.
[4] Schmidt C, Hocke T, Denkena B. Deep learning-based classification of production
defects in automated-fiber-placement processes. Prod Eng Res Devel 2019;13:501–9.
https://doi.org/10.1007/s11740-019-00893-4.
[5] Harik R, Saidy C, Williams SJ, Gurdal Z, Grimsley B. Automated Fiber Placement
Defect Identity Cards: Cause, Anticipation, Existence, Significance, and Progression.
SAMPE 2018 - Long Beach, CA, Long Beach,CA,U.S.A.: 2018.
[6] Denkena B, Schmidt C, Weber P. Automated Fiber Placement Head for Manufacturing of
Innovative Aerospace Stiffening Structures. Procedia Manufacturing 2016;6:96–104.
https://doi.org/10.1016/j.promfg.2016.11.013.
[7] Denkena B, Schmidt C, Timmermann M, Friedel A. An optical-flow-based monitoring
method for measuring translational motion in infrared-thermographic images of AFP
processes. Prod Eng Res Devel 2021. https://doi.org/10.1007/s11740-021-01084-w.
[8] Object Research Systems (ORS) Inc. Dragonfly 2021.
http://www.theobjects.com/dragonfly.
[9] Friedel A, Heimbs S, Horst P, Schmidt C, Timmermann M. Representative structural
element approach for assessing the mechanical properties of automated fibre placement-
induced defects. Composites Part C: Open Access (Submitted for Review) 2022.
[10] GOM Metrology. GOM Correlate Pro 2020. https://www.gom.com/en/products/gom-
suite/gom-correlate-pro.
[11] Deuschle HM. 3D failure analysis of UD fibre reinforced composites: Puck’s theory
within FEA. Stuttgart: Inst. für Statik und Dynamik der Luft- und
Raumfahrtkonstruktionen; 2010.
[12] Croft K, Lessard L, Pasini D, Hojjati M, Chen J, Yousefpour A. Experimental study
of the effect of automated fiber placement induced defects on performance of composite
laminates. Composites Part A: Applied Science and Manufacturing 2011;42:484–91.
https://doi.org/10.1016/j.compositesa.2011.01.007.
... Similar study was carried out to analyse the compression performance by Friedel. A et al. 19 Figure 1. Modified ATP system with new components shown within the boxes that are magnified on the right side to improve steering control of thin tape placement and also shown below separately. ...
Conference Paper
Composite structures offer higher lightweight performance when compared to traditional metals yet lack sufficient tolerance for damage onset. The focus of this effort is on thin-ply thermoset prepreg materials manufactured by Hexcel, which have shown higher resistance to damage initiation under applied load when compared to standard ply-thickness materials. To manufacture large structures currently Automated Tape placement (ATP) process is used. However, composites manufactured using thin-ply materials require placement of many more layers to achieve the same net volume of coverage. Gaps and overlaps between adjacent prepreg layers are unavoidable with tape placement and can result in degradation of mechanical properties due to these changes in the microstructure. Placement of thin-ply tape will also cause microstructural variability due to the low handling stiffness of this material. Modifications to existing automated tape placement equipment were made to address the steering of the thin-ply tapes which resulted in improved placement accuracy. Unidirectional and quasi-isotropic panels were fabricated from 12´´ wide unidirectional prepreg sheet with traditional hand layup method and compared with panels fabricated with ATP process using a ¼´´ unidirectional prepreg tape from the same 12´´ prepreg stock under varying processing conditions. All panels were cured in the autoclave with the recommended cycle. This paper compares the effects of processing parameters on the mechanical properties and the microstructure of the processed panels. This was achieved by characterizing the microstructure using C-scans, confocal microscopy and conducting mechanical tests.
... Similar study was carried out to analyse the compression performance by Friedel. A et al. 19 Figure 1. Modified ATP system with new components shown within the boxes that are magnified on the right side to improve steering control of thin tape placement and also shown below separately. ...
Article
Manufacturing of large composite structures is labor-intensive and challenging. Traditional processing methods can be replaced by automated approaches such as Automated Tape Placement (ATP). The focal point of this effort is on thin-ply thermoset prepreg materials which have shown better resistance to damage tolerance under applied stress when compared to standard ply-thickness materials in recent times. However, composites processed by thin-ply materials require more placement iterations to achieve the same net volume of coverage as the standard ply-thickness materials. This can increase the probability of placement defects such as gaps and overlaps, which may degrade the mechanical performance of the composite. Modifications to the existing ATP system are made to address the issues of placement defects, which resulted in improved accuracy for thin ply placement. Unidirectional and quasi-isotropic panels are fabricated from 12’’ wide unidirectional prepreg sheet with the conventional hand layup (CHL) method for the baseline property comparison with panels fabricated by the ATP process using a ¼’’ wide slit tape. This paper investigates the effects of processing defects and material variability on the mechanical properties such as tensile, Open hole tension and Bearing tests of thin ply composites. In addition, the microstructure of the fabricated thin-ply composite panels is analysed using ultrasonic C-scans and confocal microscopy to examine the processing and material quality.
Article
Full-text available
In this paper, a 3D finite element modelling approach is presented to assess the effects of manufacturing defects within composite structures. The mesoscale modelling approach derives the stress-strain response of a composite structure from a representative structural element. A set of tensile and bending loads is used to compute its ABD-Matrix. The boundary conditions of the model are described in detail as is the extraction of the strain and curvature response. The derived stiffness from the presented modelling approach is compared to the classical lamination theory and the models' shortcomings are discussed. Finally, the influence of a gap, an overlap and two different-sized fuzzballs on the macroscopic mechanical properties of a composite structure are evaluated using the presented multiscale modelling approach, thereby providing stiffness matrices influenced by the defects for the use in global models of composite parts.
Article
Full-text available
This paper presents a novel method for a precise localization of the automated-fiber-placement head, without the need for a data access to the machine control. It is based on a sub-pixel accurate optical-flow-algorithm which determines information about the heads movement by means of the material flow in sequences of IR images. Using local curvatures in the temperature field of the IR images, feature matrices are created which can locally be compared to the features of successive images. Thus, the translation between images become visible. This enables the possibility to perform an accurate ( 16.8\,\upmu{\mathrm{m}} 16.8 μ m ) and self-sufficient process monitoring that additionally is capable of capturing the motion and position information of the AFP system and can be linked to existing algorithms for defect detection and classification.
Conference Paper
Full-text available
Automated Fiber Placement (AFP), a major composite manufacturing process, can result in many defects during the layup process that often require manual corrective action to produce a part with acceptable quality. These defects are the main limitation of the technology and can be hard to categorize or define in many situations. This paper provides a thorough definition and classification of all AFP defects. This effort constitutes a comprehensive and extensive library relevant to AFP defects. The defects selected and defined in this work are based on understanding and experience from the manufacture and research of advanced composite structure. Proper classification of these defects required an in-depth literature review and consideration of various viewpoints ranging from designers, manufacturers, analysts, and inspection professionals. Collectively, these sources were utilized to develop the most accurate view of each of the individual defect types. The results are presented as identity cards for each defect type, intended to provide researchers and the manufacturing industry a clear understanding of the (1) cause, (2) anticipation, (3) existence, (4) significance, and (5) progression of the defined AFP defects. The link between AFP defects and process planning, layup strategies, and machining was also investigated. Categorization of all important automated fiber placement defects is presented.
Article
Full-text available
In the research project “High-performance Production of CFRP Structures” (HP CFK) a new automated fiber placement (AFP) system for laying thermoset CFRP (carbon fiber-reinforced plastic) slit tapes was developed. Its novel, modular designed laying head faces current industrial needs and challenges of prospective carbon light weight applications, e. g. future aerospace stiffening structures. Thus, its compaction unit is optimized for producing complex-curved structures. To allow approximating slopes on curved geometries, it consists of several height-adjustable rollers which, in addition, are each pressure controlled to enable an individual compacting pressure for the lay-up on materials with different compression strength (e. g. foams, metals). Furthermore, the design of the laying heads cutting unit aided the manufacturing of complex structures while being located as near as possible to the nip point to allow very short minimum placement paths. This paper introduces into the general design of the modular laying head as well as preliminary results of validation studies regarding several process limits.
Article
Automated fiber placement (AFP) is a composite manufacturing technique used to fabricate complex advanced air vehicle structures that are lightweight with superior qualities. The AFP process is intricate and complex with various phases of design, process planning, manufacturing, and inspection. An understanding of each of these phases is necessary to achieve the highest possible manufacturing quality. This literature review aims to summarize the entire AFP process from the design of the structure through inspection of the manufactured part to generate an overall understanding of the lifecycle of AFP manufacturing. The review culminates with highlighting the challenges and future directions for AFP with the goal of achieving a closed loop AFP process.
Article
This paper presents a deep learning-based approach for the detection and classification of production defects that complements an existing thermographic online monitoring system for Automated-Fiber-Placement (AFP) processes. The detection and classification procedure is performed in two stages. In the first stage, the system monitors each tow individually and classifies its process status. Furthermore, it detects and classifies production defects that affect individual tows such as a tow-twist. In the second stage, the system monitors the total width of the faultless tows. In this stage, production defects effecting multiple tows, for example gaps or overlaps, are detected and classified. Twelve different deep convolution neural networks (CNN) with three various architectures are learned supervised relating to different data sets. The performance of both identification stages is explored separately before the entire system will be set up. Therefore, the thermal images of the data sets are superimposed by noise to test the performance of the selected CNN.
Automated Fiber Placement Defects: Automated Inspection and Characterization
  • C Sacco
  • A B Radwan
  • R Harik
  • M Van Tooren
Sacco C, Radwan AB, Harik R, Van Tooren M. Automated Fiber Placement Defects: Automated Inspection and Characterization, Long Beach, CA: 2018.
3D failure analysis of UD fibre reinforced composites: Puck's theory within FEA. Stuttgart: Inst. für Statik und Dynamik der Luft-und Raumfahrtkonstruktionen
  • H M Deuschle
Deuschle HM. 3D failure analysis of UD fibre reinforced composites: Puck's theory within FEA. Stuttgart: Inst. für Statik und Dynamik der Luft-und Raumfahrtkonstruktionen; 2010.