Satyanarayana Mummana,

Unfortunately, i cannot give definite answer to this question. Because, stochastic search techniques contain randomness process. Therefore their performances can  change from problem to problem. Many parameters can effect the algorithms performances such as  problem size, number of  constraint functions and constraint function type.

If your problem is large scale, PSO algorithm can show better performance. In addition, there are many improved version GA which can be more efficient for yor problem. 

Satyanarayana Mummana,

Unfortunately, i cannot give definite answer to this question. Because, stochastic search techniques contain randomness process. Therefore their performances can  change from problem to problem. Many parameters can effect the algorithms performances such as  problem size, number of  constraint functions and constraint function type.

If your problem is large scale, PSO algorithm can show better performance. In addition, there are many improved version GA which can be more efficient for yor problem. 

 All the three (PSO, GA, ACO) are good in optimization and it depends on how you formulate your objective function

Before coding and running meta-heuristic algorithms for a specific kind of model, you can not certainly mention which one of them is more proper for the proposed model unless there are other researches that have compared them before. Moreover, these algorithms should be compared for different size of the model.

Dear Mr Mummana, 

you can find a comparative study between GA and PSO alg. in this paper:

Lo Brutto OA, Guillou SS, Thiébot J, Gualous H. Comparing Particle Swarm Optimization method and Genetic Algorithm applied to the tidal farm layout optimization problem. In: IEEE International Conference on Electrical Sciences and Technologies in Maghreb 2016 (CISTEM’16), Marrakesh, Morocco.

Best regards

Angelo

All the three (PSO, GA, ACO) are good in optimization

it depends on objective function and how you adapt the algorithm to work for the given problems

All three algorithms are good. For More details, refer the following paper:

A Survey: Swarm Intelligence vs. Genetic Algorithm by Keisam Thoiba Meetei

All three optimization techniques are better, all you need to do is to experiment first, it is also reliant on your problem

Exploration and Exploitation capability decides the performance of the algorithm. The algorithms which have good exploration and exploitation, are considered to be better one. Of course, it also depends on the given fitness function. It depends on the numbers of constant variables (parameters, factor) involved in the algorithm.