Computation on arrays using loops

58 views (last 30 days)
Kacper Witasinski on 19 Feb 2021
Edited: Kacper Witasinski on 23 Feb 2021
Hello,
I want to create code that will do operations on specified values from the arrays. I have array X and array Y, where array Y is X-dependent. Array X looks more or less like this: [0 0 0 0 0 0 0 0 0 0 0 45.9 45.98 45.99 46 46 46 46 0 0 0 0 0 0 0 ...], it represents pulse-like behaviour. I want matlab to save in workspace the values of array Y that corresponds to the every last "0" of the array X before the pulse ''46'' happens on the length of the whole array.
I expect that for/while loop has to be applied, but I am not sure which one and how the statements should look like. I would appreciate any help. Thank You. I am using matlab2020b version.

Stephen Cobeldick on 19 Feb 2021
Y = [2.659717,2.656496,2.656496,2.656173,2.662294,2.661328,2.660039,2.620416,2.614295,2.606242,2.600765,2.600443,2.590779,2.591745,2.590457,2.586913,2.58498,2.583048,2.585303,2.578216,2.577571,2.580793,2.57886,2.573061,2.575638]
Y = 1×25
2.6597 2.6565 2.6565 2.6562 2.6623 2.6613 2.6600 2.6204 2.6143 2.6062 2.6008 2.6004 2.5908 2.5917 2.5905 2.5869 2.5850 2.5830 2.5853 2.5782 2.5776 2.5808 2.5789 2.5731 2.5756
X = [0,0,0,0,0,0,0,46.24,46.24,46.24,46.24,46.24,46,46,46,46,46,45.99,45.99,45.99,45.99,45.99,46,46,46,46,46,46,46,46,46,46,46]
X = 1×33
0 0 0 0 0 0 0 46.2400 46.2400 46.2400 46.2400 46.2400 46.0000 46.0000 46.0000 46.0000 46.0000 45.9900 45.9900 45.9900 45.9900 45.9900 46.0000 46.0000 46.0000 46.0000 46.0000 46.0000 46.0000 46.0000
idx = diff(X~=0)>0;
out = Y(idx)
out = 2.6600
Kacper Witasinski on 23 Feb 2021
Works, but one note for people dealing with same problem in the future. Matlab returned an error "Dimensions of matrices being concatenated are not consistent." After I transposed idx everything works perfectly.
Regards and thank You once more Stephen.

Steven Lord on 22 Feb 2021
Since you seem to be describing change point detection, see the ischange function and/or the Find Change Points task in the Live Editor.

KALYAN ACHARJYA on 19 Feb 2021
Edited: KALYAN ACHARJYA on 19 Feb 2021
"I want matlab to save in workspace the values of array Y that corresponds to the every last "0" of the array X before the pulse ''46'' happens on the length of the whole array."
Is this? Note for this problem loop can be avoided
idx=find(diff(X)==-46)+1
This idx gives the indices of x, where x equal to zero and it's corresponding previous data to be equal to 46
Afterwards
Y=X(idx)
Kacper Witasinski on 19 Feb 2021
The values of X array given by me are really exemplary values that should give general scope of the behaviour.
Corresponding values of Y are more or less constant (the difference is in third/fourth number after the decimal point) for zeros of X array. When the pulse happen, the values are decreasing in exponential way and then values are increasing in exponential way when pulse ends. I can show You how graphs looks like to give You the idea.
So I will write once again. For majority of time the values of array X are equal to 0. At some instant, there is a pulse and the values of array Y decrease. I need the exact values of array Y that corresponds to the last zero value of array X before the pulse happen for the whole length of array (Imagine that there are 10 pulses).
Small fragment of my data:
Y= [2,659717 2,656496 2,656496 2,656173 2,662294 2,661328 2,660039 2,620416 2,614295 2,606242 2,600765 2,600443 2,590779 2,591745 2,590457 2,586913 2,58498 2,583048 2,585303 2,578216 2,577571 2,580793 2,57886 2,573061 2,575638]
X=[0 0 0 0 0 0 0 46,24 46,24 46,24 46,24 46,24 46 46 46 46 46 45,99 45,99 45,99 45,99 45,99 46 46 46 46 46 46 46 46 46 46 46 and so on]