A WE-OSELM Online Prediction Algorithm to Engine Control

Based on the time series online prediction model, this paper proposes a WE-OSELM online prediction algorithm, which can control the product life and performance in real time in the field of engine, and effectively manage huge system engineering. The prediction model is closely related to the actual application of system engineering product modules and has great operability. This paper proposes the IPW-OSELM algorithm and error compensation algorithm with information perception weight as the core point, and relies on a dual parallel single hidden layer feedforward neural network to efficiently integrate related algorithms to form a WE-OSLEM algorithm. Among them, the information perception weight clarifies the relationship between point similarity and regional similarity in the case of fluctuations in the effectiveness of the data characteristics of the prediction object or mutations in a limited range, so that the online prediction algorithm has a more acute data characteristic perception ability in the process of responding to changes in data characteristics.