Abstract: The existing harmony search algorithm (HS) in advance convergence, slow convergence, parameter selection is poor and easy to fall into local optimum, we propose an adaptive harmony search algorithm parameters improved method (DCNHS), the algorithm the new method was added and the introduction of dynamic Cauchy, normally distributed random number with adaptive parameters to achieve improved adaptive algorithm parameters. With the current performance of the more prominent of several improved HS algorithm, PSO algorithm, GA algorithm compares the improved algorithm proposed (DCNHS) in a number of standard test functions optimization algorithm simulation optimization is more accurate, more rapid convergence, and has strong ability to escape from local optima.