APIS - Acoustic process control in the sawmill

Reference number 2018-04300
Coordinator RISE Research Institutes of Sweden AB - RISE Research Institutes of Sweden AB, Borås
Funding from Vinnova SEK 500 000
Project duration November 2018 - October 2019
Status Ongoing
Venture Banbrytande idéer inom industriell utveckling
Call Banbrytande idéer inom industriell utveckling - 2018

Purpose and goal

To define useful audio data streams for an AI based vibroacoustic monitoring system in sawmills, to predict malfunctions and avoid production disturbance. After the end of the project, a number of malfunctions will be analyzed and possible to predict.

Expected results and effects

The industrial benefit is, in the long run with a fully developed system, that the downtime of the sawmill can decrease with several percent, that energy use can be optimized, work environment will improve through reduced downtime and better sound environment, and reduced raw material use through better process control. The project has the potential to strengthen the competitiveness internationally for the Swedish sawmill industry.

Planned approach and implementation

Interviews of operators, where they will describe how and where errors occur, how the errors manifest themselves, and what error states they can detect by means of sounds. Install microphones and vibration sensors in a sawmill, record sound and vibration, analyze the sound with deep learning, and draw conclusions about when to stop production before unplanned malfunctions or breakdowns occur. The data collection is an iterative process that is repeated until the machine learning system works as expected.

The project description has been provided by the project members themselves and the text has not been examined by our editors.