Swarm optimization algorithms are some of the most efficient ways to solve complexity optimization problems, especially the NP-hard problems. Wolf Search Algorithm (WSA) is a new addition to the family of swarm optimization algorithms. However there are some inherent shortcomings of these algorithms, which is the performance depends heavily on the manually chosen parameters values. One possible approach to solve this limitation is using self-adaptive methods in parameter selection. In this paper we propose some self-adaptive method for WSA to facilitate automatic selection of parameter values. Differential evolution crossover function to iteratively fine-tune the parameters values. Both random and core-guided methods are attempted to choose the evolution agents. In this paper the self-adaptive methods are implemented, certain improvement is shown from the experimental results.