Prediction of Gyro Motor´s State Based on Grey Model and BP Neural Network

The prediction accuracy of grey theory was limited by it´s high requirement of data´s smoothness. BP neural is adept in solving nonlinear problem and performs well in self-adaption and self organization, but it´s training effect and efficiency was limited by the number of data. A hybrid model combined advantages of grey theory and BP neural network is put forward based on analysis of gyro motor´s state parameters. And then, grey theory, BP neural network and the hybrid model were constructed respectively to model and predict the parameters. The results prove the validity and accuracy of hybrid model.