BFGS-GSO for global optimization problems
Aijia Ouyang, Libin Liu, Guangxue Yue, Xu Zhou, Kenli Li · Journal of Computers · 2014
To make glowworm swarm optimization (GSO) algorithm solve multi-extremum global optimization more effectively, taking into consideration the disadvantages and some unique advantages of GSO, the paper proposes a hybrid algorithm of Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm and GSO, i.e., BFGS-GSO by adding BFGS local optimization operator in it, which can solve the problems effectively such as unsatisfying solving precision, and slow convergence speed in the later period. Through the simulation of eight standard test functions, the effectiveness of the algorithm is tested and improved. It proves that the improved BFGS-GSO abounds in better multi-extremum global optimization in comparison with the basic GSO.