How to find (x,y) for peaks of a plot?

Question

Neil Solan 2018-4-9

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/393757-how-to-find-x-y-for-peaks-of-a-plot

回答： Star Strider 2018-4-9

I am using findpeaks() to get the (x,y) coordinates of the peaks of my plot, but it is giving me the x value in weird numbers. Some research has me thinking that it's giving me the x's in an index(?) but I do not know what that means and how to get it into a usable form. Thanks.

UPDATE:

I think I may understand. When I run the code, it is telling my that my locations of peaks are [180,301]. This number should be closer to 1. Is this telling me that my locations are the 180th and 301st term of the r vector? Is that what indexing means?

Code:

clear
clc
close all
%Constants:
M = 3.98*10^7; %[kg]
m = 6.6*10^5; %[kg]
k_f = 35*10^6; %[N/m]
k_d1 = 580.4*10^3; %[N/m]
k_d2 = 546.1*10^3; %[N/m]
c_d = 37.135*10^3; %[N*s/m]
F_0 = 50*10^3; %[N]
omega = 0.7:0.001:1.2; %[rad/s]
omega_1 = sqrt(k_f/M); %[rad/s]
omega_2_1 = sqrt(k_d1/m); %[rad/s]
omega_2_2 = sqrt(k_d2/m); %[rad/s]
r = omega/omega_1;
alpha1 = omega_2_1/omega_1;
alpha2 = omega_2_2/omega_1;
mu = m/M;
delta_st = F_0/k_f; %[m]
zeta = c_d/(2*m*omega_1);
%Case 1:
for i = 1:length(omega)
    X_M1(i) = F_0/(sqrt((k_f-(M*omega(i)^2))^2));
end
X_m1 = 0;
%Case 2:
for i = 1:length(omega)
    X_M2(i) = abs((alpha1^2-r(i)^2)/(((1-r(i)^2)*(alpha1^2-r(i)^2))-(mu*alpha1^2*r(i)^2)))*delta_st;
end
for i = 1:length(omega)
    X_m2(i) = abs((alpha1^2)/((1-r(i)^2)*(alpha1^2-r(i)^2)-(mu*alpha1^2*r(i)^2)))*delta_st;
end
%Case 3:
for i = 1:length(omega)
    X_M3(i) = sqrt(((alpha2^2-r(i)^2)^2+(2*zeta*r(i))^2)/((((1-r(i)^2)*(alpha2^2-r(i)^2))-(mu*alpha2^2*r(i)^2))^2+((2*zeta*r(i))*(1-r(i)^2-mu*r(i)^2))^2))*delta_st;
end
for i = 1:length(omega)
    X_m3(i) = sqrt(((alpha2^2)^2+(2*zeta*r(i))^2)/(((1-r(i)^2)*(alpha2^2-r(i)^2)-(mu*alpha2^2*r(i)^2))^2+((2*zeta*r(i))*(1-r(i)^2-mu*r(i)^2))^2))*delta_st;
end
%Case 4:
zeta_optimal = sqrt((mu*(3-(sqrt(mu/(mu+2)))))/(8*(1+mu)^3));
tuned_alpha = 1/(1+mu);
for i = 1:length(omega)
    X_M4(i) = sqrt(((tuned_alpha^2-r(i)^2)^2+(2*zeta_optimal*r(i))^2)/((((1-r(i)^2)*(tuned_alpha^2-r(i)^2))-(mu*tuned_alpha^2*r(i)^2))^2+((2*zeta_optimal*r(i))*(1-r(i)^2-mu*r(i)^2))^2))*delta_st;
end
for i = 1:length(omega)
    X_m4(i) = sqrt(((tuned_alpha^2)^2+(2*zeta_optimal*r(i))^2)/(((1-r(i)^2)*(tuned_alpha^2-r(i)^2)-(mu*tuned_alpha^2*r(i)^2))^2+((2*zeta_optimal*r(i))*(1-r(i)^2-mu*r(i)^2))^2))*delta_st;
end
subplot(2,1,1)
plot(r,X_M1,'LineWidth',2)
hold on
plot(r,X_M2,'-.r','LineWidth',2)
hold on
plot(r,X_M3,'.g','LineWidth',1)
hold on
plot(r,X_M4,'--b','LineWidth',1)
axis([0.5 1.5 0 0.05])
ylabel('X_M')
xlabel('r')
legend('Skyscraper w/ no mass damper','Skyscraper w/ undamped mass','Skyscraper w/ damped mass','Optimally Tuned')
subplot(2,1,2)
plot(r,X_m1,'LineWidth',2)
hold on
plot(r,X_m2,'-.r','LineWidth',2)
hold on
plot(r,X_m3,'.g','LineWidth',2)
hold on
plot(r,X_m4,'--b','LineWidth',1)
axis([0.5 1.5 0 0.25])
ylabel('X_m')
xlabel('r')
[pksM1,locsM1] = findpeaks(X_M1);
[pksM2,locsM2] = findpeaks(X_M2);
[pksM3,locsM3] = findpeaks(X_M3);
[pksM4,locsM4] = findpeaks(X_M4);

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Star Strider 2018-4-9

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/393757-how-to-find-x-y-for-peaks-of-a-plot#answer_314264

在 MATLAB Online 中打开

The second output are the vector indices corresponding to the identified peaks. If you want to find the ‘r’ coordinate corresponding the ‘pksM1’, for example in this findpeaks call:

[pksM1,locsM1] = findpeaks(X_M1);

it would be:

rM1 = r(locsM1);

You could then plot the peaks with a marker, for example as:

plot(rM1, pksM1, 'p')

to plot a star at each peak.