In order to solve the parameter adjustment problems of adaptive stochastic resonance system in the areas of weak signal detection, this article presents a new method to enhance the detection efficiency and availability in the system of two-dimensional Duffing based on particle swarm optimization. First, the influence of different parameters on the detection performance is analyzed respectively. The correlation between parameter adjustment and stochastic resonance effect is also discussed and converted to the problem of multi-parameter optimization. Second, the experiments including typical system and sea clutter data are conducted to verify the effect of the proposed method. Results show that the proposed method is highly effective to detect weak signal from chaotic background, and enhance the output SNR greatly.