Perhaps something like this —
T1 = readtable('https://www.mathworks.com/matlabcentral/answers/uploaded_files/994900/fantest.xlsx%20-%20Sheet2.xlsx', 'VariableNamingRule','preserve')
VN = T1.Properties.VariableNames;
t = T1.hours;
x = T1.('axis x');
y = T1.('axis y');
z = T1.('axis z');
vel = cumtrapz(t,[x y z]);
velt = sqrt(sum(vel.^2,2));
figure
subplot(4,1,1)
plot(t, vel(:,1))
grid
title('x')
ylim([-1 1]*700)
subplot(4,1,2)
plot(t, vel(:,2))
grid
title('y')
ylim([-1 1]*700)
subplot(4,1,3)
plot(t, vel(:,3))
grid
title('z')
ylim([-1 1]*700)
subplot(4,1,4)
plot(t, velt)
grid
title('Net')
ylim([-1 1]*700)
sgtitle('Velocity')
Fs = 1;
V = lowpass(T1.Voltage, 0.015, Fs);
I = lowpass(T1.Current, 0.015, Fs);
figure
plot(t, V./I)
grid
xlabel('Time')
ylabel('Z')
title('Impedance')
EDIT — (11 May 2022 at 22:36)
The ‘z’ velocity appears to be integrating a constant, so correcting for that would likely be appropriate, since it is unlikely that a constant acceleration leading to a linearly-increasing velocity is what is actually being recorded here.
I doubt that taking the Fourier transform of the velocity would be very revealing, so I did not do it here.
.
