Time series prediction based on process neural networks and its applications

To the difficulty of expression of the temporal accumulation in the time series using conventional time series prediction methods, a time series prediction method based on process neural network is proposed. Time series short-term prediction model and long-term prediction model based on the proposed method are developed respectively, and the corresponding learning algorithms are given. The effectiveness of this two models and their learning algorithms are proved by the lubricating oil iron concentration prediction in the aircraft engine condition monitoring, and the test results are satisfactory.