Improved artificial bee colony algorithm based on self-adaptive random optimization strategy

In order to effectively overcome the disadvantages of the traditional artificial bee colony (ABC) algorithm, i.e., its tendency to fall into local optima and low search speed, an improved ABC algorithm based on the self-adaptive random optimization strategy (SRABC) is proposed. First, the improved algorithm was derived from the self-adaptive method to update the new location of an ABC to improve the correlation within the bee colony. It converges swiftly and obtains the optimal solution for the benchmark function. Second, the bidirectional random optimization mechanism was used to restrain the search direction for the fitness function in order to improve the local search ability. Moreover, the particle swarm optimization algorithm regarded as the initial value of the SRABC algorithm was introduced at the initial stage of the improved ABC algorithm to increase the convergence rate, search precision and searchability, and greatly reduce the search space. Finally, simulation results for benchmark functions show that the proposed algorithm has obviously better performance regarding the search ability and convergence rate, which also prevents early maturing of algorithm.