Detecting tool wear in face milling with different workpiece materials

D. W. Cho, W. C. Choi, H. Y. Lee

    Research output: Contribution to journalConference articlepeer-review

    Abstract

    This paper proposes a neural network for the decision-making system for monitoring tool wear while working materials such as A16061, SB41, SM45C. The raw cutting forces signals are filtered and processed with adaptive AR modeling. The AR parameters and cutting conditions are used as input to the neural network along with the frequency band energy. The experimental results show that each neural network trained for each specified material can recognize tool wear with a more than 85% detection rate. When the normalized tensile strength of each material is used as additional input to the unified neural network, the network still has a success rate higher than 80%.

    Original languageEnglish
    Pages (from-to)I/-
    JournalKey Engineering Materials
    Volume183
    Publication statusPublished - 2000
    Event4th International Conference on Fracture and Strength of Solids - Pohang, South Korea
    Duration: 2000 Aug 162000 Aug 18

    ASJC Scopus subject areas

    • General Materials Science
    • Mechanics of Materials
    • Mechanical Engineering

    Fingerprint

    Dive into the research topics of 'Detecting tool wear in face milling with different workpiece materials'. Together they form a unique fingerprint.

    Cite this