Underwater metal identification method based on PSO-BP neural network

There will be an eddy current effect between the energized coil and the nearby metal conductor, and different metals have different conductivity and permeability, so that the equivalent impedance of the corresponding detector coil is different. In this paper, the capacitance and inductance parallel oscillation circuit is used to convert the change of coil equivalent impedance into voltage signal, and the neural network identification model is obtained through off-line training, so as to identify metal types. In this paper, a set of experimental device for eddy current metal type identification based on single-chip microcomputer system as the core is designed. The results show that the identification method of PSO-BP neural network and eddy current technology can effectively identify metal types, which lays a theoretical foundation for the popularization and application of intelligent metal detectors.