linear optimization using optimization toolbox and minimize array sum

I had dataset of load for 7 days, 1min time interval and I want to put a threshold value. Now claculate a difference D=(load-Threshold) and this D array should be positive by taking modulus and then a array values 'V' with corresponding to the each load data to minimize the sum of array D all data. How to minimize the sum of all elements of D and find the array values of 'V'.
For example:
a = 30;
b = 150;
D = (b-a).*rand(100,1) + a; % load data
Th=110; % threshold
Diff= abs(D-TH);
Now I want to minimize the sum of "Diff" and find a corresponding vector V(100*1) vector for all values.

4 Comments

Sorry, but I don't understand the general description of your problem.
Maybe a simple example can help to clarify.
Just take vector D(100*1) values between 50 to150 and there is one Threshold lets say 90 and I want to find out the positive differnece PD=abs(D-Th); and then minimize the sum of PD and find the values of vector V(100*1), after minimizin the sum of PD.
Thanks
If you are given D and Th, then PD=abs(D-Th) is given. Then also the sum of PD is given. How do you want to minimize the sum of PD if it is already given ?
Want vector V, for D+V-Th so the PD should be minimum

Sign in to comment.

Answers (2)

https://math.stackexchange.com/questions/113270/the-median-minimizes-the-sum-of-absolute-deviations-the-l-1-norm
Diff= abs(D-TH);
Now I want to minimize the sum of "Diff"
Threshold = median(D)

8 Comments

Th is fix for all values and after calculating the Diff, its all values should be minimized and find Vector V(100*1) such that D+V-Th=0;
Th is known in advance ? Then how can you optimize anything if everything is fixed ?
Want to find out V such that the sum of difference (D-Th) is minimum .
if we neglect that D+V-Th=0 conditon then how can find V with minimizing Diff??
Why do you neglect something that you want to achieve????
Wonder if you really know what you want.
Bcoz that constraint is by my misunderstanding and that is not a case in my original case
Thanks

Sign in to comment.

Categories

Asked:

on 31 Jul 2019

Commented:

on 31 Jul 2019

Community Treasure Hunt

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

Start Hunting!