Hybrid Glowworm Swarm Optimization And Its Applications

Glowworm swarm optimization (GSO) Algorithm is a novel swarm Intelligence search algorithm which derives from simulating the social behavior of glowworm swarm in the nature depended on fluorescein courtship or looking for companion.This algorithm has been successfully applied to the noise test of sensor, multimodal functions optimization, simulated robot, multiple signal source tracking and localization, harmful gas leaking localization, and so on. It has shown good performances. The Glowworm Swarm Optimization algorithm (GSO) gets more and more attention from people and gradually becomes a new research hot spot in the research areas of computational intelligence. Though achieved great success in its applications, the algorithm itself has some shortages, such as low precision and slow convergence.Firstly, in order to solve the problem of low precision of the Glowworm Swarm Optimization algorithm (GSO), GSO combines the fine structure of ethnic group evolution algorithm to form the glowworm swarm optimization algorithm based tribes (TGSO) which solve the function optimization problem. The experimental results show that the effect of TGSO is better than GSO. Secondly, in order to solve the problem of slow convergence of GSO, the quantity of populations is improved according to the different colors of light and the levy flights with stronger stochastic are applied into the GSO, so double glowworm swarm co-evolution optimization algorithm with levy flights (LDGSO) is presented, then the improved algorithm is applied to the problem of function optimization and multi-objective optimization. The effect of approved algorithm is better than genetic algorithm and basic GSO. The last, according to the virtues of invasive weed optimization algorithm (IWO), such as all the seeds of the invasive weed is distributed nearby their father weed by using the normal distribution. Both the overall search ability and the local search ability are improved due to the normal distribution. Moreover, in order to raise the overall performance of the IWO, survival of the fittest of the competitive strategy is adopted by the IWO, so the individual with poor adaptability in the every iteration will be cleaned out. According to the advantage of the IWO, IWO and GSO are mixed into a novel algorithm with the crossover strategy in this paper; hybrid algorithm of IWO-GSO is put forward. Then the improved algorithm is used to solve the problem of multi-objective optimization. The simulation results show that improved algorithm has greatly improved in validity and reliability.