Artificial Bee Colony (ABC) is an optimization algorithm that simulates the foraging behavior of honey bees. It is a population based search technique whose performance depends largely on the distribution of initial population. Generally, uniform distributions are preferred since they best reflect the lack of knowledge about the optimum’s location. Moreover, these are easy to generate as most of the programming languages have an inbuilt function for generating uniformly distributed random numbers. However, in case of a population dependent optimization algorithm like that of ABC, random numbers having uniform probability distribution may not be a good choice as they may not be able exploit the search space fully. This paper uses quasi random numbers based on Halton sequence for the initial distribution and have compared the simulation results with initial population generated using uniform distribution. The proposed variant, termed as Halton based ABC (H-ABC), is validated on a set of 15 standard benchmark problems, 6 nontraditional shifted benchmark functions proposed at the special session of CEC2008, and has been used for solving the real life problem of estimating the cost model parameters. Numerical results indicate the competence of the proposed algorithm.
Hybrid Artificial Bee Colony and Monarchy Butterfly Optimization Algorithm (HABC-MBOA)-based cluster head selection for WSNs
