Abstract:The BP neural network is increasingly used in the soft measurement modeling, compared with the traditional method, the BP neural network information processing can reduce the data analysis and modeling work, but there are also easy to fall into local minimum and the initial weights randomly selected defects. In order to solve the defects of traditional BP neural network, the thesis introduced in the process of data preprocessing, principal component analysis (PCA), when the input of the BP network weights is introduced into the genetic algorithm (GA), and finally achieve the purpose of make up for the BP neural network defects; Introduces in detail the process and steps of improved algorithm, the improved BP neural network was applied to the soft measurement of the aircraft hydraulic fluids detection, first analyze the aviation aircraft hydraulic fluids soft measurement parameters, including the selection of auxiliary variables and data preprocessing, and then based on the improved BP neural network modeling and simulation experiments. The experimental results show that the improved BP neural network model of the generalization ability is stronger, more widely, can achieve better measuring result, which make the BP neural Network can be used even more widely.