plot is showing blank

1 次查看(过去 30 天)
I am not able to generate the plot. It is showing empty.
Please help me to solve the same
Thank you
clear all
clc
%% calculation of alpha angle
for i=1:2
alpha(i)= (pi*22)/180;
gamma(i)=0;
beta(i)= atan(cos(alpha(i))*sin(gamma(i))./(1+sin(alpha(i))*sin(gamma(i))));
end
%% calculation of thickness
R=0.4;
t(1)= R.*cos(alpha(1))./cot(alpha(1)+ beta(1));
t(2)= R.*cos(alpha(2))./cot(alpha(2)+ beta(2));
%%
n_1=1.49;
theta_1= asin(n_1.*sin(alpha(1)))-alpha(1);
%%
T=0.2;
H_1=0.1;
t(3)= (H_1-t(1)*(tan(alpha(1))-cot(theta_1)))./(cot(theta_1)-tan(alpha(2)));
t(4)= (t(2)*(tan(alpha(2))-cot(theta_1))+ T*(tan(alpha(1))-tan(alpha(2)))-H_1)./(tan(alpha(1))-cot(theta_1));
%%
h(1)=(T-t(1)-t(4))*tan(alpha(1));
h(2)=(T-t(3)-t(2))*tan(alpha(2));
%%
n_2=1.49;
lamda_min=0.004;
lamda_max=0.007;
phase= h(1)*(n_1 -1)./(lamda_min)+ h(2)*(n_2-1)./(lamda_min);
%%
T(1)=0.1;
e1= (sin(pi*((1-(phase/2*pi))))/(pi*((1-(phase/2*pi)))))^2;
e2= (sin(pi*(t(1)/T))/(pi*(t(1)/T)))^2;
e3= (sin(pi*(t(2)/T))/(pi*(t(2)/T)))^2;
e4= (sin(pi*(t(3)/T))/(pi*(t(3)/T)))^2;
e5= (sin(pi*(t(4)/T))/(pi*(t(4)/T)))^2;
%%
% R(j)=linspace(0,0.4,101);
syms x b(x)
f=0.025;
R=0.4;
b(x) = int(R-sqrt(R.^2-x.^2),0, f);
c(x) =sqrt((1/f)*b(x));
x =linspace(0,0.025,100);
e6= exp(((-4*pi*c(x))/lamda_min).^8);
effic= e1*e2*e3*e4*e5*e6;
fplot(effic,[0 0.025]);
% fplot(effic,x);
% efficline(0, 'k')
% syms lamda
% pide=int(e1*e2*e3*e4*e5*e6, lamda_min,lamda_max);
% fpide= (1/(lamda_max-lamda_min))*pide;
  2 个评论
Kundan Prasad
Kundan Prasad 2021-12-16
exactly it is showing the same
Alberto Cuadra Lara
Alberto Cuadra Lara 2021-12-16
编辑:Alberto Cuadra Lara 2021-12-16
Hello Kundan,
If you check effic, you will see that all the values are infinitive, and this is because e6 is also infinitive. You can do the verification by transforming the symbolic variable to double
double(effic)
double(e6)
This is the reason the plot is empty.
Best,
Alberto

请先登录,再进行评论。

采纳的回答

Alberto Cuadra Lara
Alberto Cuadra Lara 2021-12-17
Hello Kundan,
Some comments:
  • H1, I could not find the value in the paper. See Ref 22. and verify the value.
  • R_T is the parametric variable in case you want to reproduce Fig. 5.
  • Check per steps the values of the variables and think if it make sense each value.
  • Compare a case in which you have final results in order to check that this script works.
Best,
Alberto.
function nu_m_mean = matlab_answers_diffraction()
% Function that computes the final difracction efficiency from Yang 2017
% Initialization
t = zeros(1, 4);
% Constants
alpha = (pi*22)/180 * ones(1, 2);
gamma = zeros(1, 2);
beta = atan((cos(alpha) .* cos(gamma)) ./ (1 + sin(alpha) .* sin(gamma)));
n1 = 1.49; % refractive index of substrate material - top
n2 = 1.49; % refractive index of substrate material - bottom
T = 100; % [micrometer]
H1 = 100; % microstructure height of double layer HDEs
lambda_min = 0.4; % [micrometer]
lambda_max = 0.7; % [micrometer]
m = 1; % diffraction order
f = 1.4; % feed rate [micrometer/r]
N = 1000;
R_T_vector = linspace(50, 400, N); % tool radius [micrometer]
% Compute Mean diffraction efficiency (PIDE)
for i = length(R_T_vector):-1:1
R_T = R_T_vector(i);
nu_m_mean(i) = compute_PIDE;
end
% Plot
plot(R_T_vector, nu_m_mean);
% NESTED FUNCTION
function nu_m_mean = compute_PIDE
% Compute Mean diffraction efficiency (PIDE)
% Diffractive angle
theta = asin(n1 * sin(alpha(1))) - alpha(1);
% Distances
t([1, 3]) = compute_distances_1_3(R_T, alpha([1,2]), beta([1,2]));
t(2) = compute_distances_2(alpha, t(1), theta, H1);
t(4) = compute_distances_4(alpha, t(3), theta, H1, T);
% Effective microstructure heights
h1 = compute_heights(alpha(1), T, t(1), t(4));
h2 = compute_heights(alpha(2), T, t(2), t(3));
% Phase delay
phi = @(lambda) compute_phase(h1, h2, n1, n2, lambda);
% Diffraction efficiency with a shadowing effect
e0 = @(lambda) compute_c(m - phi(lambda)/(2*pi));
e1 = compute_c(t(1)/T);
e2 = compute_c(t(2)/T);
e3 = compute_c(t(3)/T);
e4 = compute_c(t(4)/T);
nu_s = @(lambda) e0(lambda) * e1 * e2 * e3 * e4;
% RMS of the elements
diamond_tool_profile = @(x) R_T - sqrt(R_T^2 - x.^2);
sigma = sqrt(1/f * integral(diamond_tool_profile, 0, f));
% Mean diffraction efficiency (PIDE)
tau = @(lambda) exp((-4*pi .* sigma ./ lambda).^2);
nu_m = @(lambda) nu_s(lambda) .* tau(lambda).^4;
nu_m_mean = 1/(lambda_max - lambda_min) * integral(nu_m, lambda_min, lambda_max);
end
end
% SUB-PASS FUNCTIONS
function value = compute_distances_1_3(R_T, alpha, beta)
% Compute distances 1 or 3 (same function)
value = (R_T * cos(alpha)) ./ cot(alpha + beta);
end
function value = compute_distances_2(alpha, t, theta, H)
% Compute distance 2
value = (H - t * (tan(alpha(1)) - cot(theta))) / (cot(theta) - tan(alpha(2)));
end
function value = compute_distances_4(alpha, t, theta, H, T)
% Compute distance 4
value = (T * (tan(alpha(1)) - tan(alpha(2))) + t * (tan(alpha(2)) - cot(theta)) - H) / (tan(alpha(1)) - cot(theta));
end
function value = compute_heights(alpha, T, t1, t2)
% Compute effective microstructure heights
value = (T - t1 - t2) * tan(alpha);
end
function value = compute_phase(h1, h2, n1, n2, lambda)
% Compute phase delay
value = h1./lambda .* (n1 - 1) + h2./lambda .* (n2 - 1);
end
function value = compute_c(x)
% Compute function c from argument x
value = (sin(pi * x) / (pi * x))^2;
end
  2 个评论
Kundan Prasad
Kundan Prasad 2021-12-17
Dear Sir,
I have put the value same as mention in the refrence 22.
The attached reference pdf file is given below
file:///D:/silicon/diffractive/diffractive_3_1.pdf
Still i am getting this plot. It is totally different from actual plot.
Please have a look.
Thank you
Kundan Prasad
Kundan Prasad 2021-12-21
Dear sir,
Still I am not able to plot same as given plot. I have tried by substituting with different value.
Please have a look once sir.
Thank you

请先登录,再进行评论。

更多回答(2 个)

Cris LaPierre
Cris LaPierre 2021-12-16
Your values are all Inf. Consider checking your equations.
%% calculation of alpha angle
for i=1:2
alpha(i)= (pi*22)/180;
gamma(i)=0;
beta(i)= atan(cos(alpha(i))*sin(gamma(i))./(1+sin(alpha(i))*sin(gamma(i))));
end
%% calculation of thickness
R=0.4;
t(1)= R.*cos(alpha(1))./cot(alpha(1)+ beta(1));
t(2)= R.*cos(alpha(2))./cot(alpha(2)+ beta(2));
%%
n_1=1.49;
theta_1= asin(n_1.*sin(alpha(1)))-alpha(1);
%%
T=0.2;
H_1=0.1;
t(3)= (H_1-t(1)*(tan(alpha(1))-cot(theta_1)))./(cot(theta_1)-tan(alpha(2)));
t(4)= (t(2)*(tan(alpha(2))-cot(theta_1))+ T*(tan(alpha(1))-tan(alpha(2)))-H_1)./(tan(alpha(1))-cot(theta_1));
%%
h(1)=(T-t(1)-t(4))*tan(alpha(1));
h(2)=(T-t(3)-t(2))*tan(alpha(2));
%%
n_2=1.49;
lamda_min=0.004;
lamda_max=0.007;
phase= h(1)*(n_1 -1)./(lamda_min)+ h(2)*(n_2-1)./(lamda_min);
%%
T(1)=0.1;
e1= (sin(pi*((1-(phase/2*pi))))/(pi*((1-(phase/2*pi)))))^2;
e2= (sin(pi*(t(1)/T))/(pi*(t(1)/T)))^2;
e3= (sin(pi*(t(2)/T))/(pi*(t(2)/T)))^2;
e4= (sin(pi*(t(3)/T))/(pi*(t(3)/T)))^2;
e5= (sin(pi*(t(4)/T))/(pi*(t(4)/T)))^2;
%%
syms x b(x)
f=0.025;
R=0.4;
b(x) = int(R-sqrt(R.^2-x.^2),0, f);
c(x) =sqrt((1/f)*b(x));
x =linspace(0,0.025,100);
e6= exp(((-4*pi*c(x))/lamda_min).^8);
effic= e1*e2*e3*e4*e5*e6;
% View resulting values
double(subs(effic))
ans = 1×100
Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf Inf
  2 个评论
Kundan Prasad
Kundan Prasad 2021-12-16
Dear Sir,
the equation used in the following code are attached below

请先登录,再进行评论。


Alberto Cuadra Lara
Alberto Cuadra Lara 2021-12-16
Hello again Kundan,
I have to go and couldn't find the exact error. Some comments:
  • Check the value of the constants and units, all in micrometers?
  • H1 is less than h1. This cannot be, see Fig. 4.
  • Try to include all constants at the beginning of the script. It would be easier for you to evaluate different cases.
I have tried to solve it from scratch, hope this helps.
function nu_m = matlab_answers_diffraction()
% Function that computes the final difracction efficiency from Yang 2017
% Initialization
t = zeros(1, 4);
% Constants
alpha = (pi*22)/180 * ones(1, 2);
gamma = zeros(1, 2);
beta = atan((cos(alpha) .* cos(gamma)) ./ (1 + sin(alpha) .* sin(gamma)));
n1 = 1.49; % refractive index of substrate material - top
n2 = 1.49; % refractive index of substrate material - bottom
R_T = 10; % [micrometer]
T = 100; % [micrometer]
H1 = 50; % microstructure height of double layer HDEs
lambda_min = 0.004; % [micrometer]
lambda_max = 0.007; % [micrometer]
m = 1;
f = 0.5; % [micrometer/r]
% Diffractive angle
theta = asin(n1 * sin(alpha(1))) - alpha(1);
% Distances
t([1, 3]) = compute_distances_1_3(R_T, alpha([1,2]), beta([1,2]));
t(2) = compute_distances_2(alpha, t(1), theta, H1);
t(4) = compute_distances_4(alpha, t(3), theta, H1, T);
% Effective microstructure heights
h1 = compute_heights(alpha(1), T, t(1), t(4));
h2 = compute_heights(alpha(2), T, t(2), t(3));
% Phase delay
phi = compute_phase(h1, h2, n1, n2, lambda_min);
% Diffraction efficiency with a shadowing effect
e0 = compute_c(m - phi/(2*pi));
e1 = compute_c(t(1)/T);
e2 = compute_c(t(2)/T);
e3 = compute_c(t(3)/T);
e4 = compute_c(t(4)/T);
nu_s = e0 * e1 * e2 * e3 * e4;
% RMS of the elements
dummy_fun = @(x) R_T - sqrt(R_T^2 - x.^2);
sigma = sqrt(1/f * integral(dummy_fun, 0, f));
% Final diffraction efficiency
tau = exp((-4*pi * sigma / lambda_min)^2);
nu_m = nu_s * tau^4;
end
% SUB-PASS FUNCTIONS
function value = compute_distances_1_3(R_T, alpha, beta)
% Compute distances 1 or 3 (same function)
value = (R_T * cos(alpha)) ./ cot(alpha + beta);
end
function value = compute_distances_2(alpha, t, theta, H)
% Compute distance 2
value = (H - t * (tan(alpha(1)) - cot(theta))) / (cot(theta) - tan(alpha(2)));
end
function value = compute_distances_4(alpha, t, theta, H, T)
% Compute distance 4
value = (T * (tan(alpha(1)) - tan(alpha(2))) + t * (tan(alpha(2)) - cot(theta)) - H) / (tan(alpha(1)) - cot(theta));
end
function value = compute_heights(alpha, T, t1, t2)
% Compute effective microstructure heights
value = (T - t1 - t2) * tan(alpha);
end
function value = compute_phase(h1, h2, n1, n2, lambda)
% Compute phase delay
value = h1./lambda .* (n1 - 1) + h2./lambda .* (n2 - 1);
end
function value = compute_c(x)
% Compute function c from argument x
value = (sin(pi * (x)) / (pi * (x)))^2;
end
  1 个评论
Kundan Prasad
Kundan Prasad 2021-12-17
Dear Sir,
I have spotted the mistake i.e instead of lambda_min = 0.004; lambda_max = 0.007; it should be
lambda_min = 0.4; lambda_max = 0.7;
I want to plot same as fig.5 but i am not able to generate it.
Please have a look once.
Thank you

请先登录,再进行评论。

类别

Help CenterFile Exchange 中查找有关 Calculus 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by