Abstract: Since the teachinglearning based optimization (TLBO) algorithm was easily trapped into local optima in solving unconstrained numerical optimization problems， a teachinglearning based optimization algorithm with crossover operation was proposed. The crossover operation of differential evolution was incorporated into TLBO， this operation was effectively integrating teaching and learning stage and it was beneficial to enhance local search and balance exploitation and exploration. The numerical results show that the proposed algorithm is better than TLBO， ITLBO and other intelligent methods in terms of optimization precision， convergence speed and robustness， and the algorithm has good perspectives.

Key words: teachinglearning based optimization algorithm, local optima, crossover operation, exploitation, exploration