Wood prediction using Artificial Intelligence in Modelling (WAIMO)

Important results from the project

The purpose and aim of the project was to develop a measurement method for the diameter of the input wood and from that, then create a predictive model of how the incoming wood would affect the fiber wall thickness in the process. Within the project we have succeeded in developing a measuring method. and studied the variation of incoming wood. The project was unsuccessful in achieving a predictive model, but created the conditions for this to be possible in the future.

Expected long term effects

The work that has been conducted has made it possible to measure the dimensions of the incoming logs and given a greater understanding of the connection between wood and fiber wall thickness, as well as provided quantitative measures of how the fiber wall thickness varies over time. Both in terms of short and long term. This work has given us good conditions to work on the goal in order to be able to predict the fiber wall thickness and with the help of creating better products and more stable processes.

Approach and implementation

Installation of measuring systems took place at Iggesund´s bruk, which is part of the Holmen Group. The installation itself was done by Andritz together with local technicians. Laboratory materials were extracted from several of Holmen´s industries. Analyzes have been performed by RISE Bioekonomi & RISE SICS together with support from Holmen AB and Pulpeye. The combined results show that the concept of predicting fiber wall thickness should be possible and of value for the mills, but more work is needed to get there.