diff options
author | Matt Argao <mcargao@email.wm.edu> | 2013-03-31 22:35:49 -0400 |
---|---|---|
committer | Matt Argao <mcargao@email.wm.edu> | 2013-03-31 22:35:49 -0400 |
commit | 4d08c162d5ce4ac8b7eaeb9d985bfef62c886e99 (patch) | |
tree | 45733367770ddc4585faa09976b403a858d5d8e5 | |
parent | 2f3d161985ae2e57ed966981e58eb117d3a49c71 (diff) | |
download | mode_match-3.0.tar.gz mode_match-3.0.zip |
Added descriptions to each function.v3.0
-rw-r--r-- | energy_vs_stability.m | 25 | ||||
-rw-r--r-- | find_min.m | 4 | ||||
-rw-r--r-- | fitness.m | 125 | ||||
-rw-r--r-- | fitter_check.m | 54 | ||||
-rw-r--r-- | gaussian_focus.m | 17 | ||||
-rw-r--r-- | gbeam_propagation.m | 8 | ||||
-rw-r--r-- | gbeam_propagation_froward_only.m | 9 | ||||
-rw-r--r-- | mode_match.m | 29 | ||||
-rw-r--r-- | optics_placer.m | 1 | ||||
-rw-r--r-- | pick_visualization.m | 10 | ||||
-rw-r--r-- | remove_similar_soln.m | 4 | ||||
-rw-r--r-- | self_gbeam_propagation.m | 38 | ||||
-rw-r--r-- | solution_stability.m | 9 | ||||
-rw-r--r-- | solution_visualization.m | 64 | ||||
-rw-r--r-- | stability_visualization.m | 7 |
15 files changed, 246 insertions, 158 deletions
diff --git a/energy_vs_stability.m b/energy_vs_stability.m index 796c631..1915449 100644 --- a/energy_vs_stability.m +++ b/energy_vs_stability.m @@ -1,10 +1,22 @@ function [] = energy_vs_stability( possible_energy, stability, index, stability_max ) -%ENERGY_VS_STABILITY Summary of this function goes here -% Detailed explanation goes here +%Plots energy vs. stability +% Using data sent from mode_match, this function plots energy vs. +% stability of each solution for eyed comparisons. n = size(stability,2); - c = 1:n; + %Color code data points + if n == 3 + c = [1, 0, 0; 0, 1, 0; 0, 0, 1]; + elseif n == 2 + c = [1, 0, 0; 0, 1, 0]; + elseif n == 1 + c = [1, 0, 0]; + else + c = 1:n; + end + + %Reorganize stability to coincide with energies for i=1:n fixed_stability(i,1) = stability(i); end @@ -13,6 +25,7 @@ function [] = energy_vs_stability( possible_energy, stability, index, stability_ energy(i,1) = possible_energy(index(i)); end + %Graph scatter plot figure(n+1) textCell = arrayfun(@(x,y) sprintf('(%3.2f, %3.2f)',x,y),energy,fixed_stability,'un',0); scatter(energy,fixed_stability,25,c,'filled') @@ -23,13 +36,11 @@ function [] = energy_vs_stability( possible_energy, stability, index, stability_ dx =0.0000000005; dy =0.0000000005; + %Label data points relative to solution number for i = 1:n n_solution = num2str(i); text(energy(i)+dx, fixed_stability(i)+dy, n_solution, 'FontSize',10); end -% for i = 1:n -% text(energy(i), fixed_stability(i), textCell{i}, 'FontSize',8) -% end - + end @@ -1,7 +1,5 @@ function [ x_sol, energy ] = find_min( index, lens_placement, fitness_simplified, MaxFunEvals, MaxIter ) -%FIND_MIN Summary of this function goes here -% Detailed explanation goes here - +%Minimizes energy for given solution [x_sol, energy] = fminsearch(fitness_simplified, lens_placement(index,:), optimset('TolFun',1e-5,'MaxFunEvals',MaxFunEvals,'MaxIter', MaxIter)); end @@ -1,45 +1,16 @@ -function [Energy, Waist, Penalty] = fitness( q_0, q_final, x_final, optics_positions, optics_focal_length, lambda )
+function [Energy, Waist, Penalty] = fitness( q_0, q_final, x_final, optics_positions, optics_focal_length, lambda, self_flag )
%Outputs fitness of suggested solution
- x0 = 0;
- Energy = 0;
-
- x1=optics_positions(1);
- x2=optics_positions(2);
- x3=optics_positions(3);
-
- % penalty calculation
- % do not put lenses too close to each other and end positions
- lens_size=0.03;
-
- d(1)=abs(x1-x2);
- d(2)=abs(x2-x3);
- d(3)=abs(x1-x3);
-
- d_from_start=abs(x0-optics_positions);
- d_from_end=abs(x_final-optics_positions);
-
-
- d=cat(2, d, d_from_start, d_from_end);
-
- coef = 1;
-
- penalty_lenses_too_closely_spaced =coef*sum( exp(-(d/(lens_size)).^12) );
-
- Energy = Energy + penalty_lenses_too_closely_spaced;
-
- % make sure that lenses are between ends
- d_from_start=(x0-optics_positions);
- d_from_end=(optics_positions-x_final);
-
- d = cat(2, d_from_start, d_from_end);
-
- coef = 1;
- distance_scaling=100;
- penalty_lenses_outside_optical_path = coef * sum(1 + tanh(distance_scaling*d));
-
- Energy = Energy + penalty_lenses_outside_optical_path;
-
- % make collimated region between 2nd and 3rd lens
+x0 = 0;
+Np=20; % # of pts between start position and second lens
+N_collimated = 10; % # of pts between second lens and third lens to be collimated region
+Energy = 0;
+
+x1=optics_positions(1);
+x2=optics_positions(2);
+x3=optics_positions(3);
+
+if self_flag == 1
+ % make collimated region between 2nd and 3rd lens
[w0, r0] = q2wr(q_0, lambda);
[ w, w_pos ] = self_gbeam_propagation( w0, optics_positions, optics_focal_length, x0, lambda );
@@ -51,20 +22,84 @@ function [Energy, Waist, Penalty] = fitness( q_0, q_final, x_final, optics_posit Energy = Energy + penalty_not_collimated_beam;
-% % waist at end matches desired waist
+ % waist at end matches desired waist
coef = 10;
[waist_desired, r_desired] = q2wr(q_final, lambda);
penalty_waist = coef *((( waist_desired-w(end) )/waist_desired)^2);
Energy = Energy + penalty_waist;
-
+
coef=10;
dist_between_desired_and_final_waist_location =w_pos(end) - x_final;
penalty_final_waist_position = coef*dist_between_desired_and_final_waist_location^2;
-
+
Energy = Energy + penalty_final_waist_position;
+
+else
+ % check if waists match in forward and backward propagation
+ x_array1=linspace(x0,x2,Np);
+ x_array2=linspace(x2, x3, N_collimated);
+ x = cat(2,x_array1,x_array2);
+
+ q_f_trial_forward = gbeam_propagation(x,q_0,x0,optics_placer(optics_positions, optics_focal_length));
+ [Waist_trial_forward, Radius_trial_forward] = q2wr(q_f_trial_forward, lambda);
+ q_f_trial_backward = gbeam_propagation(x,q_final,x_final,optics_placer(optics_positions, optics_focal_length));
+ [Waist_trial_backward, Radius_trial_backward] = q2wr(q_f_trial_backward, lambda);
+
+ Penalty_waist_mismatch = sum(abs((Waist_trial_forward-Waist_trial_backward)./min(Waist_trial_forward, Waist_trial_backward)))/Np;
+
+ Energy = 1e-2*Penalty_waist_mismatch;
+
+ %intialize intermediate points between lenses
+ q_intermediate = q_f_trial_forward((x2<x) & (x<x3));
+ lambda_over_waist_sq = (-imag(1./q_intermediate)); %with numerical factor
+
+ coef = 1e-3;
+ penalty_not_collimated_beam = coef * sum(exp((std(lambda_over_waist_sq)/mean(lambda_over_waist_sq)/abs(optics_positions(2) - optics_positions(3)))));
+ Energy = Energy + penalty_not_collimated_beam;
+
+
+end
+
+
+
+% penalty calculation
+% do not put lenses too close to each other and end positions
+lens_size=0.03;
+
+
+
+d(1)=abs(x1-x2);
+d(2)=abs(x2-x3);
+d(3)=abs(x1-x3);
+
+d_from_start=abs(x0-optics_positions);
+d_from_end=abs(x_final-optics_positions);
+
+
+
+d=cat(2, d, d_from_start, d_from_end);
+
+coef = 1;
+
+penalty_lenses_too_closely_spaced =coef*sum( exp(-(d/(lens_size)).^12) );
+
+Energy = Energy + penalty_lenses_too_closely_spaced;
+
+% make sure that lenses are between ends
+d_from_start=(x0-optics_positions);
+d_from_end=(optics_positions-x_final);
+
+d = cat(2, d_from_start, d_from_end);
+
+coef = 1e-2;
+distance_scaling=100;
+penalty_lenses_outside_optical_path = coef * sum(1 + tanh(distance_scaling*d));
+
+Energy = Energy + penalty_lenses_outside_optical_path;
+
+Penalty = [ penalty_lenses_too_closely_spaced; penalty_lenses_too_closely_spaced; penalty_not_collimated_beam];
- Penalty = [ penalty_lenses_too_closely_spaced; penalty_lenses_outside_optical_path; penalty_not_collimated_beam; penalty_waist; penalty_final_waist_position];
end
diff --git a/fitter_check.m b/fitter_check.m index 9491295..b7866ae 100644 --- a/fitter_check.m +++ b/fitter_check.m @@ -1,39 +1,39 @@ %Permute all possible lens combinations out of set of lenses -lens_set = [.075, .203, .05, .03]; -lens_set = [.075,.203]; -lens_permutations = pick(lens_set,3,'or'); +lens_set = [.075, .203]; %Given lenses of unique focal lengths +lens_permutations = pick(lens_set,3,'or'); %3 lens solutions %Pre-defined Constants -lambda= 1.064E-6 ; -Ltot= 1.010675025828971 ; -r0= 1.0E+100 ; -w0= 2.563E-5 ; -x0= 0 ; -wf= 3.709E-5 ; -rf= 1.0E+100 ; -xf= Ltot; -q0=wr2q(w0,r0,lambda); -qf=wr2q(wf,rf,lambda); +lambda= 1.064E-6 ; %Wavelength of beam +Ltot= 1.010675025828971 ; %Length of optical system +r0= 1.0E+100 ; %Initial radius of curvature +w0= 2.563E-5 ; %Initial waist +x0= 0 ; %Starting position of beam +wf= 3.709E-5 ; %Desired final waist +rf= 1.0E+100 ; %Desired final radius +lens_width = .03; %Lens width +show_lens_width = 1; %Set to 1 to enable display of lens width on solution propagation plot +show_lens_position = 1; %Set to 1 to enable display of position of center of lens on solution propagation plot +display_prop = [show_lens_width, show_lens_position]; +n_truncate = 3; %number of digits in truncated solution +n_visualizations = 2; %number of best solutions to visualize +n_hist = 1000; %number of sample points in histogram +stability_max = 1; %max stability (y-axis) shown on energy vs. stability graph +self_flag = 0; %Set to 1 to use Self's gaussian beam propagation, otherwise set to 0 %End list +q0=wr2q(w0,r0,lambda); %Calculate intial q +qf=wr2q(wf,rf,lambda); %Calculate final q + %Mode match -[ possible_lens_placement, possible_lens_set, possible_sample_energy] = mode_match( q0, qf, Ltot, lambda, lens_permutations ); +[ possible_lens_placement, initial_lens_placement, possible_lens_set, possible_sample_energy] = mode_match( q0, qf, Ltot, lambda, lens_permutations, lens_width, self_flag ); %Remove similar solutions -n_truncate = 3; %number of digits in truncated solution [ possible_lens_placement_uniq, possible_lens_placement, possible_sample_energy, possible_lens_set, index ] = remove_similar_soln( possible_sample_energy, possible_lens_placement, possible_lens_set, n_truncate ); -n_visualizations = 5; %number of best solutions to visualize -pick_visualization( possible_sample_energy, possible_lens_placement_uniq, possible_lens_placement, possible_lens_set, index, n_visualizations, q0, qf, Ltot, lambda ); - -%Visualize fitness function for fixed f2 and f3 -lens_set = [.075, .075, .203]; -f2= 0.40361319425309 ; -f3= 0.80361319425309 ; -%fitness_simplified=@(x) fitness(q0, qf, Ltot, [x, f2, f3], lens_set, lambda ); -%figure(6) -%ezplot(fitness_simplified, [0,Ltot]) +%Visualize solutions +pick_visualization( possible_sample_energy, possible_lens_placement_uniq, possible_lens_placement, possible_lens_set, index, n_visualizations, q0, qf, Ltot, lambda, lens_width, display_prop ); -[ w, w_pos ] = self_gbeam_propagation( w0, possible_lens_placement, possible_lens_set, x0, lambda ) -%solution_visualization(q0,x0,qf,xf, optics_placer(possible_lens_placement, possible_lens_set), lambda) +%Plot energy vs. stability for each solution +[stability] = stability_visualization( possible_lens_placement_uniq, q0, qf, Ltot, possible_lens_placement, possible_lens_set, lambda, n_visualizations, n_hist, index, self_flag ); +energy_vs_stability( possible_sample_energy, stability, index, stability_max) diff --git a/gaussian_focus.m b/gaussian_focus.m index ff444ee..cceeda0 100644 --- a/gaussian_focus.m +++ b/gaussian_focus.m @@ -1,9 +1,14 @@ function [ w, s ] = gaussian_focus( w0, s0, f, lambda ) -%GAUSSIAN_FOCUS Summary of this function goes here -% Detailed explanation goes here +%Focusing of Gaussian beams based on Sidney A. Self's "Focusing of +%spherical Gaussian beams" Applied Optics, Vol. 22, Issue 5, pp. 658-661 (1983) +% Input: lambda = wavelength; +% w0 = waist before the lens; +% s0 = distance of waist before lens (s>0 before lens) +% f = focal length of lens (f>0 for converging lens) +% Output: w = new waist; +% s = position of waist from lens -zR = pi*w0^2/lambda; -s = f*(1+(s0/f-1)/((s0/f-1)^2+(zR/f)^2)); +zR = pi*w0^2/lambda; %Rayleigh Range +s = f*(1+(s0/f-1)/((s0/f-1)^2+(zR/f)^2)); %Eq. (9b) w = w0/sqrt((1-s0/f)^2+(zR/f)^2); -end - +end
\ No newline at end of file diff --git a/gbeam_propagation.m b/gbeam_propagation.m index d81f70e..585048d 100644 --- a/gbeam_propagation.m +++ b/gbeam_propagation.m @@ -1,4 +1,4 @@ -function q = gbeam_propagation(x_pos, q_in, x_in, optics_elements) +function [q] = gbeam_propagation(x_pos, q_in, x_in, optics_elements) % calculate the 'q' parameter of the Gaussian beam propagating through optical % 'optics_elements' array along 'x' axis at points 'x_pos' % takes the gaussian beam with initial q_in parameter at x_in @@ -8,7 +8,7 @@ function q = gbeam_propagation(x_pos, q_in, x_in, optics_elements) if any(x_pos >= x_in) % Forward propagation to the right of x_in - q(x_pos >= x_in) = gbeam_propagation_froward_only(x_pos(x_pos>=x_in), q_in, x_in, optics_elements); + [q(x_pos >= x_in)] = gbeam_propagation_froward_only(x_pos(x_pos>=x_in), q_in, x_in, optics_elements); end if any(x_pos < x_in) @@ -45,8 +45,10 @@ function q = gbeam_propagation(x_pos, q_in, x_in, optics_elements) % final assignment of the backwards propagating beam % which we need to flip back q(x_pos<x_in) = fliplr(q_backw); - end + end + + end %!test diff --git a/gbeam_propagation_froward_only.m b/gbeam_propagation_froward_only.m index afc1316..2cdc1d8 100644 --- a/gbeam_propagation_froward_only.m +++ b/gbeam_propagation_froward_only.m @@ -1,4 +1,4 @@ -function q = gbeam_propagation_froward_only(x_pos, q_in, x_in, optics_elements) +function [q] = gbeam_propagation_froward_only(x_pos, q_in, x_in, optics_elements) % calculate the 'q' parameter of the Gaussian beam propagating through optical % 'optics_elements' only in the positive direction along 'x' axis at points 'x_pos' % takes the gaussian beam with initial q_in parameter at x_in @@ -8,6 +8,10 @@ function q = gbeam_propagation_froward_only(x_pos, q_in, x_in, optics_elements) % % all x_pos must be to the right of x_in % x_pos must be monotonic! + + %Initialize q_lens (q at position of lens) + q_lens = zeros(1,size(optics_elements,2)); + if (any(x_pos < x_in)) error('all beam positions must be to the right of the x_in'); end @@ -45,8 +49,7 @@ function q = gbeam_propagation_froward_only(x_pos, q_in, x_in, optics_elements) index = (x_last_calc <= x_pos); x=x_pos(index); q(index)= q_after_free_space(q_last_calc,x-x_last_calc); - - + end diff --git a/mode_match.m b/mode_match.m index 0a36332..f3f7408 100644 --- a/mode_match.m +++ b/mode_match.m @@ -1,4 +1,4 @@ -function [ possible_lens_placement, possible_lens_set, possible_sample_energy] = mode_match( q0, qf, Ltot, lambda, lens_permutations ) +function [ final_possible_lens_placement, initial_possible_lens_placement, possible_lens_set, possible_sample_energy] = mode_match( q0, qf, Ltot, lambda, lens_permutations, lens_width, self_flag ) %Shuffles lenses into random positions and stores possible solutions % Shuffles over entire lens permutation array for n_shuffles times. % Afterwards, solutions are sorted by Energy and truncated to n_truncate @@ -7,6 +7,11 @@ function [ possible_lens_placement, possible_lens_set, possible_sample_energy] = n_perms = size(lens_permutations,1); n_shuffles=20; %number of random placements of lenses +initial_MaxFunEvals = 1e8; +initial_MaxIter = 10; +final_MaxFunEvals = 1e8; +final_MaxIter = 100; + %Initialize sample arrays N = n_perms * n_shuffles; possible_lens_placement = zeros(N,3); @@ -14,25 +19,33 @@ possible_lens_set = zeros(N,3); possible_sample_energy = zeros(N,1); initial_rand_lens_placement=zeros(N,3); -lens_size = .03; % physical size of the lens - for ip = 1:n_perms f3=lens_permutations(ip,3); x3=Ltot-f3; % last lense transfer collimated region to focused spot for is = 1:n_shuffles possible_lens_set((ip-1)*n_shuffles + is,:) = lens_permutations(ip,:); - initial_rand_lens_placement_tmp = sort(lens_size+(x3-2*lens_size)*rand(1,2)); + initial_rand_lens_placement_tmp = sort(lens_width+(x3-2*lens_width)*rand(1,2)); initial_rand_lens_placement((ip-1)*n_shuffles + is,:) = [initial_rand_lens_placement_tmp, x3]; end end + + + parfor i = 1:N - - fitness_simplified=@(x) fitness(q0, qf, Ltot, x, possible_lens_set(i,:), lambda ); - [x_sol, energy]=fminsearch(fitness_simplified, initial_rand_lens_placement(i,:), optimset('TolFun',1e-5,'MaxFunEvals',1e8,'MaxIter',200)); + fitness_simplified=@(x) fitness(q0, qf, Ltot, x, possible_lens_set(i,:), lambda, self_flag ); + [ x_sol, energy ] = find_min(i, initial_rand_lens_placement, fitness_simplified, initial_MaxFunEvals, initial_MaxIter ) + + initial_possible_lens_placement(i,:) =x_sol; - possible_lens_placement(i,:) =x_sol; +end + +parfor i = 1:N + fitness_simplified=@(x) fitness(q0, qf, Ltot, x, possible_lens_set(i,:), lambda, self_flag ); + [ x_sol, energy ] = find_min(i, initial_possible_lens_placement, fitness_simplified, final_MaxFunEvals, final_MaxIter ) + + final_possible_lens_placement(i,:) =x_sol; possible_sample_energy(i) = energy; end diff --git a/optics_placer.m b/optics_placer.m index 247cc81..2407691 100644 --- a/optics_placer.m +++ b/optics_placer.m @@ -1,4 +1,5 @@ function optics = optics_placer( x ,f )
+%Places optical elements within array optics for use in other functions
x1=x(1);
x2=x(2);
x3=x(3);
diff --git a/pick_visualization.m b/pick_visualization.m index 88f1989..5a11231 100644 --- a/pick_visualization.m +++ b/pick_visualization.m @@ -1,4 +1,4 @@ -function [ ] = pick_visualization( fitness_energy, possible_lens_placement_uniq, possible_lens_placement, possible_lens_set, index, n_visualizations, q0, qf, Ltot, lambda ) +function [ ] = pick_visualization( fitness_energy, possible_lens_placement_uniq, possible_lens_placement, possible_lens_set, index, n_visualizations, q0, qf, Ltot, lambda, lens_width, display_prop ) %Picks n_visualizations of sets of data and graphs each x0 = 0; @@ -6,11 +6,11 @@ n_possible_lens_placement = min(n_visualizations,size(possible_lens_placement_un for n_graph = 1:n_possible_lens_placement figure(n_graph) - [w_final_trial, r_final_trial] = solution_visualization(q0, x0, qf, Ltot, optics_placer(possible_lens_placement(index(n_graph),:), possible_lens_set(index(n_graph),:)), lambda); + [w_final_trial, r_final_trial] = solution_visualization(q0, x0, qf, Ltot, optics_placer(possible_lens_placement(index(n_graph),:), possible_lens_set(index(n_graph),:)), lambda, lens_width, display_prop); - str1=sprintf('\n f_1 = %0.4f, x_1 = %0.4f\n',possible_lens_set(index(n_graph),1),possible_lens_placement(index(n_graph),1)); - str2=sprintf('f_2 = %0.4f, x_2 = %0.4f\n',possible_lens_set(index(n_graph),2),possible_lens_placement(index(n_graph),2)); - str3=sprintf('f_3 = %0.4f, x_3 = %0.4f\n',possible_lens_set(index(n_graph),3),possible_lens_placement(index(n_graph),3)); + str1=sprintf('\n (red) f_1 = %0.4f, x_1 = %0.4f\n',possible_lens_set(index(n_graph),1),possible_lens_placement(index(n_graph),1)); + str2=sprintf(' (green) f_2 = %0.4f, x_2 = %0.4f\n',possible_lens_set(index(n_graph),2),possible_lens_placement(index(n_graph),2)); + str3=sprintf(' (blue) f_3 = %0.4f, x_3 = %0.4f\n',possible_lens_set(index(n_graph),3),possible_lens_placement(index(n_graph),3)); tstr='Solution #'; str_w = ['w_{final}= ',num2str(w_final_trial)]; str_r = [', r_{final}= ',num2str(r_final_trial)]; diff --git a/remove_similar_soln.m b/remove_similar_soln.m index 9bea732..a106aff 100644 --- a/remove_similar_soln.m +++ b/remove_similar_soln.m @@ -1,6 +1,6 @@ function [ possible_lens_placement_uniq, possible_lens_placement, possible_sample_energy, possible_lens_set, index ] = remove_similar_soln( possible_sample_energy, possible_lens_placement, possible_lens_set, n_truncate ) -%REMOVE_SIMILAR_SOLN Summary of this function goes here -% Detailed explanation goes here +%Removes similar solutions based on n_truncate (decimal tolerance to +%determin uniqueness) %Sorting possible solution according to energy [possible_sample_energy, index] = sort(possible_sample_energy); diff --git a/self_gbeam_propagation.m b/self_gbeam_propagation.m index 10bcf05..824be5a 100644 --- a/self_gbeam_propagation.m +++ b/self_gbeam_propagation.m @@ -1,11 +1,11 @@ function [ w, w_pos ] = self_gbeam_propagation( w0, x_lens, f, x0, lambda ) -%SELF_GBEAM_PROPAGATION Summary of this function goes here -% Detailed explanation goes here +%Beam propagation based on Sidney A. Self's "Focusing of spherical Gaussian +%beams". Applied Optics, Vol. 22, Issue 5, pp. 658-661 (1983) n_lens = 3; -w = w0; -w_pos = x0; +w = w0; %Initial waist +w_pos = x0; %Initial waist position for i = 1:n_lens x_from_next_lens = x_lens(i) - w_pos(i); @@ -14,32 +14,4 @@ for i = 1:n_lens w_pos(i+1) = x_lens(i) + s_new; end -% x_pos= x_lens; -% x(1)=w_pos(1); -% waist_prop(1)=w(1); -% -% x_temp=linspace(x0 ,x_pos(1)); -% waist_prop=[waist_prop, w(1)*sqrt(1+((x_temp-w_pos(1))/(pi*w(1)^2/lambda)).^2)]; -% x=[x x_temp]; -% -% for i=1:n_lens -% x_temp=linspace(x_pos(i),x_pos(i+1)); -% waist_prop=[waist_prop, w(i)*sqrt(1+((x_temp-w_pos(i))/(pi*w(i)^2/lambda)).^2)]; -% x=[x x_temp]; -% end -% -% x_temp=linspace(x_pos(end),x_pos(end)+abs(f(end))); -% waist_prop=[waist_prop, w(end)*sqrt(1+((x_temp-w_pos(end))/(pi*w(end)^2/lambda)).^2)]; -% x=[x x_temp]; -% -% -% figure(1) -% plot(x,waist_prop,'r',x,-waist_prop,'r'); -% - - - - - -end - +end
\ No newline at end of file diff --git a/solution_stability.m b/solution_stability.m index 324621f..359335c 100644 --- a/solution_stability.m +++ b/solution_stability.m @@ -1,12 +1,13 @@ -function [ hist_h, hist_x ] = solution_stability( q_0, q_final, x_final, optics_positions, optics_focal_length, lambda, n_hist ) -%SOLUTION_STABILITY Summary of this function goes here -% Detailed explanation goes here +function [ hist_h, hist_x ] = solution_stability( q_0, q_final, x_final, optics_positions, optics_focal_length, lambda, n_hist, self_flag ) +%Plots histogram of n_hist number of sample points +%Area under plot is equal to the stability of the solution. + lens_displacement = 10^-3; x = lens_displacement*randn(n_hist,3); for i=1:n_hist xd= x(i,:); - fitness_array(i) = fitness( q_0, q_final, x_final, optics_positions + xd, optics_focal_length, lambda ); + fitness_array(i) = fitness( q_0, q_final, x_final, optics_positions + xd, optics_focal_length, lambda, self_flag ); end [hist_h, hist_x] = hist(fitness_array,100); diff --git a/solution_visualization.m b/solution_visualization.m index 2cf8ce7..2b4158c 100644 --- a/solution_visualization.m +++ b/solution_visualization.m @@ -1,11 +1,19 @@ -function [waste_at_the_end, radius_at_the_end] = solution_visualization(q0,x0, qf, xf, optics, lambda) +function [waste_at_the_end, radius_at_the_end, waist_at_lens_position] = solution_visualization(q0,x0, qf, xf, optics, lambda, lens_width, display_prop) %Propagates beam with given input parameters % Forward propagation and backward propagation are taken in order to % visualize reasonable solutions. +%Array of lens positions +n_lens = size(optics,2); + +for i = 1:n_lens + lens_position(i) = optics{i}.x; +end + x=linspace(x0,xf,1000); % we will calculate beam profile between x0 and xf + %Forward propagation -q_forward=gbeam_propagation(x,q0,x0,optics); +q_forward =gbeam_propagation(x,q0,x0,optics); [w_forward,r_forward]=q2wr(q_forward, lambda); %Backward propagation @@ -13,13 +21,53 @@ q_backward=gbeam_propagation(x,qf,xf,optics); [w_backward,r_backward]=q2wr(q_backward, lambda); %Plot beam profile -plot ( ... - x,w_forward, '-r', ... - x,-w_forward, '-r', ... - x, w_backward, '-.b', ... - x, -w_backward, '-.b') -legend({'forward propagation', '', 'backward propagation', ''}) +subplot(2,1,1); plot ( ... + x,w_forward, '-r', ... + x,-w_forward, '-r', ... + x, w_backward, '-.b', ... + x, -w_backward, '-.b') +legend({'forward propagation', '', 'backward propagation', ''}) + +%Find q and waist at lens positions +waist_at_lens_position = zeros(1,n_lens); +q_lens = zeros(1,n_lens); + +for i = 1:n_lens + q_lens(i) = gbeam_propagation(lens_position(i),q0,x0,optics); +end + + +for i = 1:n_lens + waist_at_lens_position(i) = q2wr(q_lens(i), lambda); +end + + +%Plot lenses +color = ['m' 'g' 'c']; + +if display_prop(1) == 1 + for i = 1:n_lens + half_width = lens_width/2; + corrected_lens_position = lens_position(i)-half_width; + corrected_waist = waist_at_lens_position(i)*2; + rectangle('Position', [corrected_lens_position,-waist_at_lens_position(i),lens_width,corrected_waist], 'EdgeColor', color(i), 'LineWidth',1); + end +end + +if display_prop(2) == 1 + for i = 1:n_lens + x1 = optics{i}.x; + y1 = waist_at_lens_position(i); + x2 = x1; + y2 = -waist_at_lens_position(i); + + line([x1;x2], [y1;y2], 'LineWidth', 3, 'Color', color(i)); + end +end + [waste_at_the_end,radius_at_the_end] = q2wr(q_forward(end), lambda); + + diff --git a/stability_visualization.m b/stability_visualization.m index ba0a8de..37e9aad 100644 --- a/stability_visualization.m +++ b/stability_visualization.m @@ -1,12 +1,11 @@ -function [stability] = stability_visualization( possible_lens_placement_uniq, q0, qf, xf, optics_position, optics_set, lambda, n, n_hist, index ) -%STABILITY_VISUALIZATION Summary of this function goes here -% Detailed explanation goes here +function [stability] = stability_visualization( possible_lens_placement_uniq, q0, qf, xf, optics_position, optics_set, lambda, n, n_hist, index, self_flag ) +%Visualizes stability graph n_possible_lens_placement = min(n, size(possible_lens_placement_uniq,1)); for i=1:n figure(i) - [hist_h, hist_x] = solution_stability( q0, qf, xf, optics_position(index(i),:), optics_set(index(i),:), lambda, n_hist ); + [hist_h, hist_x] = solution_stability( q0, qf, xf, optics_position(index(i),:), optics_set(index(i),:), lambda, n_hist, self_flag ); subplot(2,1,2); plot(hist_x, hist_h); area = trapz(hist_x, hist_h); str_n = ['# of test points = ', num2str(n_hist)]; |