The on-line tool condition monitoring is demanded to detect the tool wear and to ensure the hole drilling process of printed circuit boards (PCB) goes on smoothly. However, due to the impact of ambient noise caused by the limited size of small drill and the laminated material of PCB, the tool wear signal features are too weak to extract. The stochastic resonance (SR) method has been proven to be effective in enhancing weak signals among various weak signal extraction. In this paper, an adaptive multistable stochastic resonance is presented to improve performance of the SR method and process the tool wear signals for PCB drilling. The differential evolution (DE) algorithm is applied to adaptively optimize potential parameters and compensation factor, which makes the SR method suitable for high frequency signal. Moreover, tool wear experiments with different drill wear are carried out to verify the effectiveness of the proposed method. The results indicate that the proposed method improves the signal-to-noise ratio and has great potential in enhancing weak signals for small drill condition monitoring in PCB drilling process.