1 算法介绍
Optimization is an essential tool to minimize or maximize functions, obtaining optimal results on costs, mass, energy, gains, among others. Actual problems may be multimodal, nonlinear, and discontinuous and may not be minimized by classical analytical methods that depend on the gradient. In this context, there are metaheuristic algorithms inspired by natural phenomena to optimize real engineering problems. No algorithm is the worst or the best, but more efficient for a given problem. Thus, a new nature-inspired algorithm called Lichtenberg Optimization Algorithm (LA) is applied in this study to solve a complex inverse damage identification problem in mechanical structures built by composite material. To verify the performance of the new algorithm, both LA and Finite Element Method (FEM) were used to identify delamination damage and the results were compared to other algorithms such as Genetic Algorithm (GA) and SunFlower Optimization (SFO). LA was shown to be a powerful damage identification tool since it was able to detect damage even in particular situations like noisy response and low damage severity.
2 部分代码
% LICHTENBERG ALGORITHM (LA) FOR NONLINEAR UNCONSTRAINED % OPTIMIZATION % % % Copyright (c) 2020, Jo茫o Luiz Pereira and Guilherme Ferreira Gomes % All rights reserved. % % % Please cite this algorithm as: % % Pereira, J.L.J., Chuman, M., Cunha Jr, S.S. and Gomes, G.F. (2020), % "Lichtenberg optimization algorithm applied to crack tip identification % in thin plate-like structures", Engineering Computations, %% Initialization clear all close all clc format long tic set(0,'DefaultAxesFontName', 'Times New Roman') set(0,'DefaultAxesFontSize', 14) set(0,'DefaultTextFontname', 'Times New Roman') set(0,'DefaultTextFontSize', 26) % Bounds UB = [5 5]; %upper bounds LB = [-5 -5]; %lower bounds d = length(LB); %problem dimension % LA Parameters pop = 20; %Population size n_iter = 100; %Max number os iterations Np = 100000; %Number of Particles (If 3D, better more than 10000) S_c = 1; %Stick Probability: Percentage of particles that can don麓t stuck in the %cluster. Between 0 and 1. Near 0 there are more aggregate, the density of %cluster is bigger and difusity is low. Near 1 is the opposite. Rc = 150; %Creation Radius (if 3D, better be less than 80, untill 150) refinement = 0.2; %use after running the algorithm once. M = 0; %M = 1 creates new Lichtenberg figures in each iteration, M=0 use same figure (fast convergence); x0 = (LB+UB)/2; %Initial Point (only for first population) disp_plot = 1; %if = 1, display LA figure for 2d problems. %Some objective test functions Fun1 = @(x) 100*(x(2)-x(1)^2)^2+(1-x(1))^2; % rosenbrok (1,1) Fun2 = @(x) (1.5-x(1)*(1-x(2)))^2+(2.25-x(1)*(1-x(2)^2))^2; % beale's (3,0.5) Fun3 = @(x) x(1)^2 +x(2)^2 +x(3)^2; % Schwefel (0,0,0) Fun4 = @(x) 100*(x(2)-x(1)^2)^2+(1-x(1))^2 + 100*(x(3)-x(2)^2)^2+(1-x(2))^2; % rosenbrok (1,1,1) Fun = Fun1; [x,fval,iter,state,population]=LA_optimization(Fun,d,pop,x0,Np,S_c,Rc,M,LB,UB,refinement,n_iter,disp_plot); % Convergence view figure, semilogy(state(:,d+2)) xlabel('Iterations') ylabel('Objective') toc %end of the program
3 仿真结果
4 参考文献
[1] Pereira J , MB Francisco, Jr S , et al. A powerful Lichtenberg Optimization Algorithm: A damage identification case study[J]. Engineering Applications of Artificial Intelligence, 2021, 97:104055.