Abstract: Coal blending characteristics have an important influence on coal blending combustion, in order to investigate the influence of coal blending scheme on coal quality characteristics and power blending characteristics of coal blending, a nonlinear analysis method is used to analyze and predict the relevant mechanical characteristics of Haragou coal processing plant under different blending schemes. Firstly, the coal quality prediction model of blended coal is established by modifying the neural network in the nonlinear method to a generalized neural network. The predicted heat Q shows that the accuracy of the prediction using the neural network can be greatly improved compared with the linearly weighted results. The application of nonlinear analysis method in coal blending problem is better, which can provide ideas and theoretical basis for the application of nonlinear problems in other aspects.