134. Atakulreka, A. and Sutivong, D., 2007, December. Avoiding local minima in feedforward neural networks by simultaneous learning. In Australasian Joint Conference on Artificial Intelligence (pp. 100-109). Springer, Berlin, Heidelberg.
Nawi, N.M., Khan, A. and Rehman, M.Z., 2013, June. A new back-propagation neural network optimized with cuckoo search algorithm. In International conference on computational science and its applications (pp. 413-426). Springer, Berlin, Heidelberg.
136. N. M. Nawi, R. Ransing, and M. Ransing. An improved conjugate gradient based learning algorithm for back propagation neural networks. International Journal of Computational Intelligence. 2007, 4, 46-55.
137. Tran-Ngoc, H., Khatir, S., De Roeck, G., Bui-Tien, T. and Wahab, M.A., 2019. An efficient artificial neural network for damage detection in bridges and beam-like structures by improving training parameters using cuckoo search algorithm. Engineering Structures, 199, p.109637.
138. Khatir, S., Tiachacht, S., Thanh, C.L., Bui, T.Q. and Wahab, M.A., 2019. Damage assessment in composite laminates using ANNPSO-IGA and Cornwell indicator. Composite Structures, 230, p.111509.
139. Rajendra, M., Jena, P.C. and Raheman, H., 2009. Prediction of optimized pretreatment process parameters for biodiesel production using ANN and GA. Fuel, 88(5), pp.868-875.
140. Yazdanmehr, M., Anijdan, S.M. and Bahrami, A., 2009. Using GA–ANN algorithm to optimize soft magnetic properties of nanocrystalline mechanically alloyed Fe–Si powders. Computational Materials Science, 44(4), pp.1218-1221.
141. Azadeh, A., Mianaei, H.S., Asadzadeh, S.M., Saberi, M. and Sheikhalishahi, M., 2015. A flexible ANN-GA-multivariate algorithm for assessment and optimization of machinery productivity in complex production units. Journal of Manufacturing Systems, 35, pp.46-75.
142. Javidrad, F., Nazari, M. and Javidrad, H.R., 2018. Optimum stacking sequence design of laminates using a hybrid PSO-SA method. Composite Structures, 185, pp.607-618.
Có thể bạn quan tâm!
- Cập Nhật Mô Hình Và Xác Định Hư Hỏng Trong Kết Cấu
- Kết Quả Phát Hiện Hư Hỏng Của Pso, Ann Và Annpso
- L. F. F. Miguel, R. H. Lopez, And L. F. F. Miguel. A Hybrid Approach For Damage Detection Of Structures Under Operational Conditions. Journal Of Sound And Vibration. 2013, 332, 4241-4260.
- Chẩn đoán dầm cầu bằng phương pháp phân tích dao động trên mô hình số hoá kết cấu được cập nhật sử dụng thuật toán tối ưu hoá bầy đàn kết hợp mạng nơ ron nhân tạo - 17
- Chẩn đoán dầm cầu bằng phương pháp phân tích dao động trên mô hình số hoá kết cấu được cập nhật sử dụng thuật toán tối ưu hoá bầy đàn kết hợp mạng nơ ron nhân tạo - 18
- Chẩn đoán dầm cầu bằng phương pháp phân tích dao động trên mô hình số hoá kết cấu được cập nhật sử dụng thuật toán tối ưu hoá bầy đàn kết hợp mạng nơ ron nhân tạo - 19
Xem toàn bộ 154 trang tài liệu này.
143. Tran-Ngoc, H., Khatir, S., Le-Xuan, T., De Roeck, G., Bui-Tien, T. and Wahab, M.A., 2020. A novel machine-learning based on the global search techniques using
vectorized data for damage detection in structures. International Journal of Engineering Science, 157, p.103376.
144. Choi, F.C., Li, J., Samali, B. and Crews, K., 2007. An experimental study on damage detection of structures using a timber beam. Journal of mechanical science and technology, 21(6), pp.903-907.
PHỤ LỤC
Code MATLAB được NCS lập trình và áp dụng trong luận án.
Dầm giản đơn
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc clear close all
%% Intinilization yt=0;
yb=0; b=0;
Code=1; t=0;
Line=0;
h = (1143+180)*1E-3;
% Node Nodes=Nodes_Sources_concrete(Code); E= 3E10;
nu=0.3;
rho=2450;%khoi luong rieng;
% Element types -> {EltTypID EltName} Types= {1 'beam';
2 'link'};
A = 0.604; % 714025E-6; % m2
Izz = 0.157; % 144730000000E-12; % m2 yb = 872E-3; % m2
yt = h-yb;
b = 558E-3; % be rong bau dam
% Sections=[SecID A ky kz Ixx Iyy Izz yt yb zt zb] Sections=[1 A inf inf 0 0 Izz yt yb b/2 b/2];
% Elements=[EltID TypID SecID MatID n1 n2 n3] Elements=Elements_Sources_concrete(Code); Elements_Sources=Elements_Sources_concrete(Code);
%%%%%%%%%%%%%%%%%%%%%%%%%% MAX DAMAGE AND MODE%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%
========================================================================
%
Max_Damage_Percent=0.5; % Max Damage Percent
nMode=10; % Get number of frequency you want (max = 25) Select_case = 2; % to show the case
%
========================================================================
% Element_Target.ID=[1;2;3;4;5;6;7;8;9;10;11;12;13;14;15;16;17;18;19;20]; Element_Target.ID=[1;2;3;4;5;6;7;8;9;10;11;12];
Element_Target.Material = 2.*ones(length(Element_Target.ID),1);
% Size of Template Element_Target.Size=size(Element_Target.ID,1); MaTrix_Case=length(Element_Target.ID); MaTrix_Case=1:1:MaTrix_Case;
for iCase =1:Select_case All_Case(iCase).Case=nchoosek(MaTrix_Case,iCase); % Creat All_Case(iCase).Size=size(All_Case(iCase).Case);
end
for iCase = 1 : Select_case
% Rows
for Rows = 1 : All_Case(iCase).Size(:,1)
% Columns
for Columns = 1 : All_Case(iCase).Size(:,2) No=All_Case(iCase).Case(Rows,Columns); All_Case(iCase).Elements(Rows,Columns)=Element_Target.ID(No,1); All_Case(iCase).Materials(Rows,Columns)=Element_Target.Material(No,1);
end end
end
% Run and plot to excel for iCase = 1:Select_case
disp(['========================== No.Case ' num2str(iCase)... ' ==========================']);
Element_Matrix=All_Case(iCase).Elements; Line=0;
Frequency_Information_ANN=0;
% Column
for Rows = 1 : All_Case(iCase).Size(:,1) Elements=Elements_Sources; No=All_Case(iCase).Case(Rows,:); Element_find=0;
for k = 1: iCase
% Element_find=zeros(1,iCase);
% Find in Sources and take that number of line
Element_find(1,k)= find(Elements(:,1) == All_Case(iCase).Elements(Rows,k));
% Material : Choosed Element_Damage.Material(1,k)=All_Case(iCase).Materials(Rows,k);
% Change Material was choosed nFind=size(Element_find,2);
for j=1:nFind Elements(Element_find(1,j),4)=Element_Damage.Material(1,j); end
end
%% Damage Loop
for Damage_Loop=0.01:0.01:Max_Damage_Percent t=t+1;
Line=Line+1;
if t>(Max_Damage_Percent*100) t=1;
end
% Damage percent Element_Information_ANN(Line,1)=t;
% Element Damage
for Next=2:size(No,2)+1
Element_Information_ANN(Line,Next)=No(1,Next-1); end
% Damage
Damage=E-E*Damage_Loop;
% Materials=[MatID E nu]; Materials= [1 E nu rho Inf
2 Damage nu rho Inf];
% Get all degree of freedom DOF=getdof(Elements,Types);
% Select degree of freedom
seldof=[0.03; 0.04; 0.05; 1.01;1.02; 25.02];
% Remove degree of freedom DOF=removedof(DOF,seldof);
% assumble in one column matrix of Stiffness and Mass [K,M]=asmkm(Nodes,Elements,Types,Sections,Materials,DOF);
% Eigen of element [phi,omega]=eigfem(K,M,nMode);
% Display eigenfrequenties a=omega/2/pi; idx=(1:1:nMode); Excel_Size=size(idx,2);
% Creat information for sheet 1 (freequency) for j=1:Excel_Size
fre_FEM=a(idx); Frequency_Information_ANN(Line,j)=fre_FEM(j,1)'; end
disp(['No.Element =' num2str(Element_Information_ANN(Line,2:end))... ' ***** Damping percent =' num2str(t) '%'...
' **** Line =' num2str(Line)]); end % Damage Loop
end % Rows
% Creat name for file of each case switch iCase
case 1 Excel_Name='Case1_beam_11.xlsx'; case 2 Excel_Name='Case2_beam_2.xlsx'; case 3
Excel_Name='Case3.xlsx'; case 4 Excel_Name='Case4.xlsx'; case 5 Excel_Name='Case5.xlsx'; case 6 Excel_Name='Case6.xlsx'; case 7 Excel_Name='Case7.xlsx'; case 8 Excel_Name='Case8.xlsx'; case 9 Excel_Name='Case9.xlsx'; case 10 Excel_Name='Case10.xlsx';
end
%Remove all row have at least one zeros Element_Information_ANN = Element_Information_ANN...
(all(Element_Information_ANN,2),:);
% Get information for write Frequency_Information_ANN; Element_Information_ANN;
%%%%%%%%%%%%%%%%%%%%%%%%%% WRITE TO EXCEL
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
xlswrite(Excel_Name,Frequency_Information_ANN,1,'A1'); xlswrite(Excel_Name,Element_Information_ANN,2,'A1'); end
disp('**********************YOUR DATA IS READY**************************');
Tấm Composite clearvars;colordef white;clf;clc close all
% materials thickness=0.2;h=thickness;kapa=0.84; rho=1;I=thickness^3/12;
% symbolic computation
syms phi pi %Create symbolic variables and functions
% liew material e2=1;e1=40*e2;g23=0.5*e2;g13=0.6*e2;g12=g13;
miu12=0.25;miu21=miu12*e2/e1;factor=1-miu12*miu21;
% angles for laminate alfas=[0,pi/2,0];% 3 layers
% upper and lower coordinates
z(1)=-(h/2);z(2)=-(h/2)+h/3;z(3)=-z(2);z(4)=-z(1);
% [Q] in 0º orientation
% transformation matrix T=[cos(phi)^2,sin(phi)^2,-sin(2*phi),0,0;...
sin(phi)^2,cos(phi)^2,sin(2*phi),0,0;...
sin(phi)*cos(phi),-sin(phi)*cos(phi),cos(phi)^2-sin(phi)^2,0,0;... 0,0,0,cos(phi),sin(phi);...
0,0,0,-sin(phi),cos(phi)];
% [Q] in structural axes qBarra=T*qbarra*T.'; QQbarra=zeros(size(alfas,2),5,5); for s=1:size(alfas,2)
for i=1:5 for j=1:5
QQbarra(i,j,s)=subs(qBarra(i,j,1),phi,alfas(s)); end
end Qbarra=double(QQbarra);
end