To improve cutting tool wear condition monitoring accuracy,wear condition monitoring method based on artificial bee colony-BP neutral network is proposed. Force sensor and vibration sensor are used to design monitoring platform. Matching pursuit algorithm is used to denoise for the signal. Characteristic parameters are extracted from time domain,frequency domain,and time-frequency domain. These characteristic parameters are reduced dimensionally by kernel principal component analysis,and fifteen parameters are ensured at last. Cutting tool wear condition recognition method based on artificial bee colony-BP neutral network algorithm is presented, and artificial bee colony algorithm is used to optimize neutral network parameters. The experimental result proves that the recognition accuracy is 78. 75% by BP neutral network and 100% by optimized BP neutral network algorithm respectively.