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,


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!

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


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 - 16

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

Xem toàn bộ nội dung bài viết ᛨ

..... Xem trang tiếp theo?
⇦ Trang trước - Trang tiếp theo ⇨

Ngày đăng: 14/07/2022