Using Particle Swarm Optimisation for high dimensionality multivariable optimisation

Question

Craig 2014-5-17

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/130001-using-particle-swarm-optimisation-for-high-dimensionality-multivariable-optimisation

编辑： Craig 2014-5-17

I'm using Pramit Biswas's implementation of Particle Swarm Optimisation , but the documentation is quite hard to understand.

I have three questions:

First, how should I pass the cost function to the function seeing as the cost function takes in 675 variables? Second, how suitable is this optimisation method suitable for high dimensionality problems like this one? (675 variables)

Finally, How many "birds" or "particles" would be a reasonable amount? I have guessed 20 is a reasonable amount, as that is what this paper suggests.

My best attempt so far is:

particle_Swarm_Optimization(20,675,constraints,@Cost2,'min',2,2,2,0.4,0.9,5000)

Where the arguments are:

function [ optimised_parameters ] = Particle_Swarm_Optimization (Bird_in_swarm, Number_of_quality_in_Bird, MinMaxRange, Food_availability, availability_type, velocity_clamping_factor, cognitive_constant, social_constant, Min_Inertia_weight, Max_Inertia_weight, max_iteration)

Is this correct? The function hasn't given any error message, and its on 14% after about 25 minutes, but I don't want to wait if its going to give me a useless result.

Thanks!