The teaching-learning-based optimizer (TLBO) algorithm is a powerful and efficient optimization algorithm. However it is prone to getting stuck in local optima. In order to improve the global optimization performance of TLBO, this study proposes a modified version of TLBO, called teaching-learning-studying-based optimizer (TLSBO). The proposed enhancement is based on adding a new strategy to TLBO, named studying strategy, in which each member uses the information from another randomly selected individual for improving its position. TLSBO is then used for solving different standard real-parameter benchmark functions and also various types of nonlinear optimal power flow (OPF) problems, whose results prove that TLSBO has faster convergence, higher quality for final optimal solution, and more power for escaping from convergence to local optima compared to original TLBO.