Fault diagnosis of underwater vehicle with neural network

In order to aim at the character that the uncertainties of the complex system of underwater vehicle (UV) result in model the system very difficult, a wavelet neural network (WNN) is proposed to construct the motion model of UV. The adjustment of the scale factor and shift factor of wavelet and weights of WNN is discussed. The WNN has the ability not only to approach the whole figure of a function but also to catch detail changes of the function, which makes the approaching effect wonderful. Residuals are achieved by comparing the output of WNN with the sensor output. Fault detection rules are distilled from the residuals to execute thruster fault diagnosis. The feasibility of the method presented is verified by simulation experiment results.