Abstract:In view of the incubation process which is highly nonlinear, large delay and time-varying and strong coupling, a combination forecasting method based on gray prediction and associative memory neural networks was proposed. Firstly, the gray prediction model and associative memory neural networks were respectively used on prediction for incubator. Then the two sub-models with weighted integration by using the variance covariance combination forecasting method was combined. The value of the temperature and humidity of incubator could be more accurate and effective. Finally, the results showed the mean square deviation of combination forecasting model was 0.9%.