Grinding Wheel Condition Monitoring using Acoustic Airborne Signals and Machine Learning

Autor: Gharaei, Ali, Wahl, Tobias, Ostad Ali Akbari, Vahid, Kuffa, Michal, Wegener, Konrad
Jazyk: angličtina
Rok vydání: 2022
Předmět:
DOI: 10.3929/ethz-b-000604806
Popis: Wear state of grinding tools has a significant influence on the grinding result. Depending on the tool wear mechanism, the grains’ cutting edges deteriorate which is referred to as grain dulling. When it happens to many active abrasive grains, large wear flats appear on the grinding tool surface, referred to as glazing. Under such circumstances, the contact area and frictional interactions between the abrasive grains and the workpiece increase. This leads to rising temperatures to high levels in the area of contact and heat accumulation whereby adhesion and chemical reaction between two surfaces are enhanced while material removal efficiency and overall grinding performance are undermined. In high-performance dry grinding processes that are often force-controlled and dressing-free, such grinding states may occasionally persist without tool self-resharpening. This paper introduces a methodology to identify the grinding process states in an automated way. A face grinding experiment is conducted with a resin-bonded corundum wheel. It has been observed that airborne acoustic signals can effectively be used to identify the glazing states of grinding tools. The solution supports process control, the performance evaluation of various grinding wheels, and an in-depth investigation of wear mechanisms.
Databáze: OpenAIRE