Prolonging the network life cycle is one of the major goals to be achieved in the wireless sensor network (WSN), and topology control is of great significance for it. To solve the problem of cluster head selection in WSN, a multi-objective cluster head selection algorithm, which is based on NSGA-II, is proposed. Many factors such as network communication distance, energy consumption, load balancing and node survival time are considered in the algorithm. The optimal number of cluster heads is determined by theoretical calculation to guide population initialization, and the orthogonal experiment mechanism is introduced to reduce the times of search and improve the efficiency of optimizing. Experiments show that the proposed algorithm, compared with the low-energy adaptive clustering hierarchy (LEACH) algorithm, has evenly distributed cluster head, balanced load and extended network life cycle. In addition, it is of higher searching efficiency than the standard NSGA-II algorithm.