Find in Library

Acoustic emission on-line detection method was used to obtain the acoustic emission information of the tank bottom. Combined with the parameters of the tank itself, a genetic algorithm(GA) and BP neural network were combined to establish a model to evaluate the corrosion of the tank. BP neural network evaluation model and GA-BP network evaluation model were established respectively, and the results of the two models were compared. The results show that the GA-BP corrosion evaluation model of the tank bottom has better adaptability and higher prediction accuracy than the BP neural network evaluation model alone. The experiment has reference value for the on-line detection of atmospheric tank corrosion, and provides support for China′s independent research and development of acoustic emission detection system of atmospheric pressure metal storage tanks.