Adaptive Adjustment of Grading Machine Settings

Authors

Markus K. Sandomeer, Jochen Köhler and Michael H. Faber

Abstract

The current European standard for machine strength grading of structural timber EN 14081 provides two different methods for the control of grading machine settings. These are either machine controlled or output controlled; however, both methods do not facilitate the efficient use of information which is gathered prior and during the runtime of the grading machine [2]. In addition, the mentioned control methods are broadly considered to be much too complex and ambiguous. Throughout Europe there are only a few experts who know how to deal with the required procedures for approval of grading machines which are addressed in the standard. As a consequence different projects and activities have recently been initiated to investigate how EN 14081 could be reorganized:
- There are extensive discussions within COST Action E53 about the existing procedures provided by EN 14081, especially with regard to the identification of common growth areas and the development of efficient control procedures for grading machine settings.
- Participants of a large pan-European project (GRADEWOOD) investigate both: the comparability of data bases of different European countries containing data of machine graded and destructively tested structural timber, and the development of alternative approaches for machine grading procedures.
- EN 14081 is currently taken under revision by the responsible experts of the Technical Committee CEN TC124/WG2.
All of these investigations can be considered to contribute to future simplifications and modifications of the relevant standard for grading of structural timber – EN 14081.
In that sense, this paper introduces the probabilistic output control method as an alternative framework for modelling the grading of structural timber. The developed model is expected to meet the demand for efficient control procedures applicable on the European situation of timber production. For this purpose flexible and fast adjustable grading models are required that utilize all available information gathered previous and during the grading procedure. Since the investigations prescribed in the following chapters have recently been discussed by the international experts of CIB-W18 (meeting 41st, 24-28 August 2008) the interested reader is Cost Action E53 Conference 29th – 30th October 2008 in Delft, The Netherlands referred to the paper presented at this occasion [4] to find a more detailed description of the methods and experimental results.

Published in/by

Proceedings COST E53 Conference End User's Needs for Wood Material and Products, Delft, The Netherlands, October 29-30, 2008.