hello sir I am working on enhancing the stability of voltage using GA. I have 3 objective function : f1- minimize power loss f2-minimize ( max L-index) f3-minimize voltage deviation by suming all entire purpose performances and the equality and inequality limitation , we get final objective function (F) . Do we consider "F" as this multi objective function? And can we calculated L-index directly using matlab?
Hello, i want to create a colour scale into a 2-dimensional Pareto Front Diagramm. Is that possible? Or is it just possible in a 3 dimensionale diagramm? Thanks :)
Hi Son, thanks for the video. I had a question related to the function file. I noticed that if one mentions output as 'Output = [F2 F1 F3]' or '... [F3 F1 F2]' instead of '... [F1 F2 F3]', the set of solutions that we get varies a lot. Could you please enlighten what the underlying reasons for that could be? Does the sequence of mentioning objective functions make any difference as to which objective function/s gets higher or lower priority? Secondly, how do we pick up the best solution from the solution set? Please let me know, based on your experience, if there is any thumb rule for the same?
Thanks! it can be related to decision variables ? I increased the population size; the same problem, still only one point. I am sure of the objective function.
How can we set the increments for the upper and lower bounds? For example, here lb=[-3 -3], ub=[3 3], and the generation of populations is chosen randomly within these bounds. So, some combinations of points x and y may be left out. How can we indicate that we want to consider x and y from -3 to 3 with a step=1, that is x=[-3:1:3] or y=linspace (-3, 3, 7) ?
@@SolvingOptimizationProblems Yes, it generates random combinations, and I guess because of this, not all possible combinations can be taken into account. What do you think, based on your experience, is it possible to find a Pareto front in Matlab from loaded combinations of points of object functions?
Thanks a lot. But two question! First, is it possible to increase the number of input variables to 3 or more? Second, is this method called NSGAII bult in algorithm from Matlab to solve optimization?
Hi, 1) yes, it is possible to increase the number of variables. 2) This is multi objective GA solver in Matlab. This is not NSGAII. For NSGA II check other videos on my channel. Thanks for watching and have a great day!
Hi, for problem formulation, objective function is the thing we want to optimize and the format of objective function in matlab is as shown in the video.
Hi, First o f all, thank you very mush for your very well explained videos. i have two questions: How the Pareto Surf points are chosen? and how could we choose the one optimal solution from the last figure Can you show me how to construct a custom plot function inside those three files (where to put it) in order to see the variation of the optimization variable at each generation thanks in advance
Hi, thanks for watching. We can't choose the pareto front. Pareto front are optimal solutions determined by the problem. Optimal solutions are determined by GA solver in this video. We should not manipulate the plot; it is determined by the problem and optimization algorithm. If you want to change the plot, please change the problem parameters. Hope it helps.
@@SolvingOptimizationProblems thanks for your quick response i think we could change something for the plot since there is a custom plot function choice , where we can code a function called at each itteration in order to see the evolution of all parameter at each population evaluation and not at each generation and for the solution choice, i know that all the pareto front are optimal solution, so are you saying that the final choice belong to us since there is more then one optimal solution?
Ah yes, now I understand what you meant. Yes, we can plot the evolution of the optimal solutions along with iterations. But we need to modified the code. Yes, we can select the final solution(s).
@@SolvingOptimizationProblems your video is awsome. i have the same question with this friend. how to plot the evolution of the optimal solutions along with iterations, i.e. the 3d pareto front in every iteration. could you please explain how to modified the code?
Hi, Its showing Invalid solver specified. Provide a solver name or handle (such as 'fmincon' or @fminunc).Type DOC OPTIMOPTIONS for a list of solvers. Error in multi_objective_genetic_algorithm_solver (line 2) options = optimoptions('gamultiobj'); Error in Main (line 14) [x,fval,exitflag,output,population,score] = multi_objective_genetic_algorithm_solver(nvars,lb,ub,PopulationSize_Data,MaxGenerations_Data,MaxStallGenerations_Data,FunctionTolerance_Data,ConstraintTolerance_Data); It is not running.
Hi its very helpful, thanks for sharing informations. Can you solve smart grid problems through this technique, it might be the same as this one. If yes, I will share equations from the article, cost and emission reduction, availability of renewable energy sources. These are the problems. Thanks in advance
@@SolvingOptimizationProblems Its ok, thank you. Can I share equations, that equations includes multiple summation bounds, therefore, I am not able to add that in this particular program. Can you please help me? I can share those equations here, or learn with panda (your site), or even on linkedin (optimization page or your personal linkedin account). Thanks
Thank you for your sharing! Can we use this technique to find the Pareto front for multi-objective optimisation when objectives are subject to the ODE system?
Is it possible to solve the two-level programming model like this..or is it different in the solution?
Sorry, I am not sure. I have never done that before.
how can we optmal allocate the DG using GWO algorithm?
I don't have background to fully understand your problems
Thank you. How to display x an y optimal values for the objective function?
Please see in the command space!
Please help me
I want to use a multiobjective programming (type Maximizes)
All results negative
Thanks
Please convert the max problem to min problem
hello sir
I am working on enhancing the stability of voltage using GA.
I have 3 objective function :
f1- minimize power loss
f2-minimize ( max L-index)
f3-minimize voltage deviation
by suming all entire purpose performances and the equality and inequality limitation , we get final objective function (F) .
Do we consider "F" as this multi objective function?
And can we calculated L-index directly using matlab?
In that case (F), you use weighted sum method - the easy but not effective method for multi-objective optimization.
How to plot graphs for ga....
This graph is generated the toolbox in Matlab
Hello, i want to create a colour scale into a 2-dimensional Pareto Front Diagramm. Is that possible? Or is it just possible in a 3 dimensionale diagramm?
Thanks :)
Yes, it is possible to apply to 2 objective function optimization
need to know that to use gamultiobj solver we have to convert our all objective functions into minimization problem
just add a coefficient to the objective function
Hi Son, thanks for the video. I had a question related to the function file. I noticed that if one mentions output as 'Output = [F2 F1 F3]' or '... [F3 F1 F2]' instead of '... [F1 F2 F3]', the set of solutions that we get varies a lot. Could you please enlighten what the underlying reasons for that could be? Does the sequence of mentioning objective functions make any difference as to which objective function/s gets higher or lower priority?
Secondly, how do we pick up the best solution from the solution set? Please let me know, based on your experience, if there is any thumb rule for the same?
1) Yes, the sequence is very importance; 2) all solutions on the pareto front are equally optimal; meaning can pick any of them
Hello, please I got one point on pareto front. Can you tell me why. Thanks!
Please check the objective function format or increase the population size.
Thanks!
it can be related to decision variables ?
I increased the population size; the same problem, still only one point.
I am sure of the objective function.
Maybe, check the upper and lower bounds of variables and/or constraints
Thanks !
Can we use this method to optimize problem with 3 variables or more?
Yes, it is possible to solve problems with more than 3 variables
How can we set the increments for the upper and lower bounds? For example, here lb=[-3 -3], ub=[3 3], and the generation of populations is chosen randomly within these bounds. So, some combinations of points x and y may be left out.
How can we indicate that we want to consider x and y from -3 to 3 with a step=1, that is x=[-3:1:3] or y=linspace (-3, 3, 7) ?
I think the solve generates the random value between with the lower and upper bounds. I can't fixed the increments
@@SolvingOptimizationProblems Yes, it generates random combinations, and I guess because of this, not all possible combinations can be taken into account. What do you think, based on your experience, is it possible to find a Pareto front in Matlab from loaded combinations of points of object functions?
Hello sir,your matlab code link doesn't work.Please give me this code link in comment section.
Hello, I don't have the code. This is GA solver in Matlab. You will get this solver if you install matlab version 2016a.
is it possible to generate Parento fronts
Hi, the figure is showing the Parento front
Which version of matlab used for this code?
2016 version of matlab.
@@SolvingOptimizationProblems 2016a or 2016b
It's 2016a. Thanks
Sir I used weighted sum method to form the objective function. How can we find the weight for each objective function?
Hello, it depends on the characteristics of the problem and opinion of decision makers.
How i can liked pso with my program code m-file in matlab
I think you can make the PSO as a function (having input and output) and then call the PSO from your program. Hope it's clear enough!
Hello sir can I use the code to optimize 4 responses each one have 4 variables ( multi-response optimization)? thanks in advance
Yes, it is possible
Thank you for your channel sir. It's very useful
So nice of you. Thanks for watching!
hi sir and thank you for this extraordinary video, but I have a problem with this code, it is not working in my Matlab, can I contact you please ?
Please send me an email.
how can I contact you regarding the matlab code?
You can contact me by email
@@SolvingOptimizationProblems does the code compatible with Matlab 2016 b because I have not version 2016 a ?
Thanks a lot.
But two question!
First, is it possible to increase the number of input variables to 3 or more?
Second, is this method called NSGAII bult in algorithm from Matlab to solve optimization?
Hi, 1) yes, it is possible to increase the number of variables. 2) This is multi objective GA solver in Matlab. This is not NSGAII. For NSGA II check other videos on my channel. Thanks for watching and have a great day!
@@SolvingOptimizationProblems best regards
@@SolvingOptimizationProblems Thanks for the answer. Does Matlab have NSGA II built-in algorithm at all?
No, Matlab does not have NSGA II
@@SolvingOptimizationProblems Which one has better performance NSGA II or GA MATLAB algorithem?
Sir please how to solve GA for function with ode45 solver
Sorry, I don't know the ode45 solver
Hi, thank you for your sharing! I want all variables to be an integers, what should I do?
Thx
Can I use round() directly in nsga2? (e.g When the function (non_domination_sort_mod) generates the next generation)
Rounding can be used here.
Hello sir , How do we make/ form the objective function for the given problem. Is there any specific procedure to make one?
Hi, for problem formulation, objective function is the thing we want to optimize and the format of objective function in matlab is as shown in the video.
@@SolvingOptimizationProblems thankyou
And can we use Matlab optimization tool for triple obj. Problems?
Is this same as multi-objective?
Yes, they are the same.
@@SolvingOptimizationProblems thankyou so much 🙏
Hii sir,
how to solve triple objective optimization with constraints?
Great question! Noted. Will try it
@@SolvingOptimizationProblems Thanks Sir
Hi,
First o f all, thank you very mush for your very well explained videos.
i have two questions:
How the Pareto Surf points are chosen? and how could we choose the one optimal solution from the last figure
Can you show me how to construct a custom plot function inside those three files (where to put it) in order to see the variation of the optimization variable at each generation
thanks in advance
Hi, thanks for watching. We can't choose the pareto front. Pareto front are optimal solutions determined by the problem. Optimal solutions are determined by GA solver in this video. We should not manipulate the plot; it is determined by the problem and optimization algorithm. If you want to change the plot, please change the problem parameters. Hope it helps.
@@SolvingOptimizationProblems thanks for your quick response
i think we could change something for the plot since there is a custom plot function choice , where we can code a function called at each itteration in order to see the evolution of all parameter at each population evaluation and not at each generation
and for the solution choice, i know that all the pareto front are optimal solution, so are you saying that the final choice belong to us since there is more then one optimal solution?
Ah yes, now I understand what you meant. Yes, we can plot the evolution of the optimal solutions along with iterations. But we need to modified the code. Yes, we can select the final solution(s).
@@SolvingOptimizationProblems your video is awsome. i have the same question with this friend. how to plot the evolution of the optimal solutions along with iterations, i.e. the 3d pareto front in every iteration. could you please explain how to modified the code?
Hi, Its showing
Invalid solver specified. Provide a solver name or handle (such as 'fmincon' or @fminunc).Type DOC OPTIMOPTIONS for a list of solvers.
Error in multi_objective_genetic_algorithm_solver (line 2)
options = optimoptions('gamultiobj');
Error in Main (line 14)
[x,fval,exitflag,output,population,score] =
multi_objective_genetic_algorithm_solver(nvars,lb,ub,PopulationSize_Data,MaxGenerations_Data,MaxStallGenerations_Data,FunctionTolerance_Data,ConstraintTolerance_Data);
It is not running.
That is because your matlab does not have that solver. Please use matlab version 2016a
How to convert maximize function to minimise function
Hello, we can use a simple formular like YY = 1/Y if Y is always positive
Can't we just change the sign
Please.. i need code for call m-file in matlab
Hi, check the link in the description. Thanks for watching!
@@SolvingOptimizationProblems thank you.. can i call m-file using run
No, it does not make any sense
Thanks for your sharing!
Many thanks for watching!
Interested please share code..
Hi, this is optimization solver in Matlab
Hi
its very helpful, thanks for sharing informations.
Can you solve smart grid problems through this technique, it might be the same as this one.
If yes, I will share equations from the article, cost and emission reduction, availability of renewable energy sources.
These are the problems.
Thanks in advance
Hi, I think it is totally possible.
@@SolvingOptimizationProblems
Its ok, thank you.
Can I share equations, that equations includes multiple summation bounds, therefore, I am not able to add that in this particular program.
Can you please help me?
I can share those equations here, or learn with panda (your site), or even on linkedin (optimization page or your personal linkedin account).
Thanks
You can send me an email on my blog. Thank you!
@@SolvingOptimizationProblems its ok.
Thank you
hi this programme shows many errors and not run
Hi, what was your error? Please keep in mind that the version of matlab used here is 2016a
@@SolvingOptimizationProblems thanks
Im ned solve bilevel programming
Noted. Many thanks for your suggestion!
👍🏼👍🏼
The idea of this video came from your suggestion. Do you remember that?
@@SolvingOptimizationProblems Yes , good work done 👍🏼
Thank you for your sharing!
Can we use this technique to find the Pareto front for multi-objective optimisation when objectives are subject to the ODE system?
Yes you can!