Modeling and Simulation Comparison of Real-time Monitoring of Particleboard Hot-Pressing Thermal Conductivity

In this paper we present a new measure method for real-time monitor of thermal conductivity of the hot-pressing process applied to particleboard. In the particleboard hot-pressing process, the monitoring data of thermal conductivity was modeled and forecasted based on BP neural network and ELM algorithm. The trained network can both accurately harmonize different sets of experimental data and precisely predict the dynamic indexes of thermal conductivity. Simulation results show that ELM prediction method not only overcome the shortcoming of the traditional real-time monitoring mechanism in precision but also avoid the problems of depending on a large number of data with high quality in existing BP network learning.