ICF10B

Figure 2: The loading scheme of the beam specimen with eccentric crack. ( ) ( ) ( ) ( ) ( ) ( ( ) ( ) ( ) ( ) ) ( ) ( ∑ ∑ ∑ ∑ = = − − − = = − − − η ω ω − − − − = η ω ω − − − − = N i n j i j i i i i i N i n j i j i i i i i t t k k H t t K t K F t K t t k k H t t K t K F t K 1 1 1 2 1 1 (1) IIS (1) IIS II 1 1 1 1 1 1 (1) IS (1) IS I 2 1 sin sin ) (12) where denotes the circular frequencies of natural oscillations of the beam without support; mean the number of modes of natural oscillations, which are taken into account in analysis; ωi 1 2 , n n       η = ω ∑η = = n i i i i i K 1 1 (1) IS 2 0 1 1 , i K uI       ω ∑η = η = = n i i i i i i K u K 1 2 (1) IIS 2 II 0 2 1 ; (1) IS K (1) IIS K IIi Ii , K K is the і u i0 th component of displacement vector for the point of load’ application; , are the static SIF for the tension of the specimen by force, uniformly distributed in the volume; are the SIF in the specimen, deformed by ith normed mode of autonomous oscillations. While calculating ( K , , the volume loading was substituted by equally distributed loading applied to the top and bottom part of specimen. For this the FEA approach was used. The mesh quality was verified by comparison with the results, obtained in [14]. The results in the range 1) IS (1) IIS K 4 6 ≤γ≤ , 0.3 0.7; eW 1 λ ε ≤ ≤ = = can be described by equations (with error less than 1.5%): ( ) ( [ )( )] ( )( ) [ ] . 0.7705 0.1051 0.3122 0.2024 0.4365 0.262 0.0861 , 1.0861 0.0606 2.3095 0.0333 1.9329 1.5083 6.2731 1.3461 4.1651 2 2 (1) IIS 2 3 2 2 1 IS BW K L l BW K L l γ − λ λ+ − λ+λγ − =λ γ − λ + λ + λ+ λ− λ +γ λ− + = − (13) The dimensionless coefficients 1 ,i 2i η η and eigenfrequencies ρ ω =ωa E i i * do not depend on the specimen dimensions. For 1 eW ε= = , 0.7 0.3≤λ≤ , 4.0 6.0 ≤ γ ≤ (here, aW LW λ = γ = ; ) and ν=0,3 using the Lancosh blocks method we found the circular frequencies of natural oscillations and corresponding to them eigenvectors. Analyzing the relations , we concluded that it is enough to chose and for evaluation of and ( ) 1 1,2 =j 1 ∑η = = n i ji 5 1 = n 6 2 = n ( )t IK ( ) K t II correspondingly. Processing the numerical data using the least square method, following polynomial approximations were proposed for ω : * i ( )( ) ( )( ( )( ) ( )( ), 4.2913 0.2151 0.2123 0.5026 0.7079 6.4583 1.2866 5.6316 , 1.509 0.104 1.2356 1.1247 5.6163 0.6612 6.0468 , 0.0096 1.7487 0.0096 0.3022 0.7018 1.0621 4.0988 0.8996 2.8588 , 1.4143 0.105 0.0763 1.1088 1.4685 0.2935 4.617 6.2351 2 2 * 4 2 * 3 2 2 2 * 2 2 2 * 1 γ − λ λ+ − λ −λγ λ− + ω = γ − λ − λ −λγ λ− + ω = γ γ− − λ − λ+ λ −λγ λ− + ω = γ − λ λ+ − λ −λγ + λ− ω = ) (14) ( )( ) ( )( ). 3.8153 0.2075 0.1916 0.8715 1.2239 1.6014 10.416 13.535 , 4.1461 0.2309 0.1549 0.7007 0.8957 10.092 1.5634 8.5572 2 2 * 6 2 2 * 5 γ − λ λ+ − λ −λγ λ− + ω = γ − λ λ+ − λ −λγ λ− + ω = The approximations (14) describe the numerical data with the deviation less, than 2%. For coefficients i1η following dependencies are proposed:

RkJQdWJsaXNoZXIy MjM0NDE=