DE EN

Wood Materal Science & Engineering  10/2024

Predicting wood moisture classes by sound frequency spectra and Explainable Machine Learning during milling

Authors:

Mehieddine Derbas, Timothy Mark Young, Stephan Frömel-Frybort, Hans- Christian Möhring & Martin Riegler

Contents:
  •          The moisture content in wood has a strong impact on the optimal process parameters during milling.
  •          In this work, wood milling was monitored by a novel optical microphone to distinguish between moisture classes (dried, conditioned, wet and frozen).
  •          Machine learning was applied to the selected spectral data to determine both the moisture classes and milling speed. The best model achieved an accuracy of 97.2%. 
View all Publications