THRESHOLD ADAPTIVE MODELING FOR DRILLING PROCESS IDENTIFICATION AND MONITORING.

Steven Hsin-Yi Lai, S. H. Hsieh

Research output: Contribution to conferencePaper

Abstract

This paper presents an adaptive modeling technique for the identification and monitoring of drilling processes. The major dynamics of drill failure are captured by a threshold mixed model using the in-process drilling vibration data. The threshold learning model and the initial guess parameters are evaluated from the first drilling hole signal. The parameters are updated by an adaptive scheme in the subsequent operations. Two control indices, the transient slope and the steady-state model residuals, are estimated. The process is monitored using a forecasting control chart generated from the control parameters. The monitoring result shows that the threshold approach is capable of predicting the wear state of a tool for various machining operations with a high variability of dynamic data.

Original languageEnglish
Pages89-94
Number of pages6
Publication statusPublished - 1987 Dec 1

Fingerprint

Drilling
Monitoring
Vibrations (mechanical)
Machining
Wear of materials
Control charts

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

@conference{64f6d4b3cbd941ab848878b5f83ded88,
title = "THRESHOLD ADAPTIVE MODELING FOR DRILLING PROCESS IDENTIFICATION AND MONITORING.",
abstract = "This paper presents an adaptive modeling technique for the identification and monitoring of drilling processes. The major dynamics of drill failure are captured by a threshold mixed model using the in-process drilling vibration data. The threshold learning model and the initial guess parameters are evaluated from the first drilling hole signal. The parameters are updated by an adaptive scheme in the subsequent operations. Two control indices, the transient slope and the steady-state model residuals, are estimated. The process is monitored using a forecasting control chart generated from the control parameters. The monitoring result shows that the threshold approach is capable of predicting the wear state of a tool for various machining operations with a high variability of dynamic data.",
author = "Lai, {Steven Hsin-Yi} and Hsieh, {S. H.}",
year = "1987",
month = "12",
day = "1",
language = "English",
pages = "89--94",

}

THRESHOLD ADAPTIVE MODELING FOR DRILLING PROCESS IDENTIFICATION AND MONITORING. / Lai, Steven Hsin-Yi; Hsieh, S. H.

1987. 89-94.

Research output: Contribution to conferencePaper

TY - CONF

T1 - THRESHOLD ADAPTIVE MODELING FOR DRILLING PROCESS IDENTIFICATION AND MONITORING.

AU - Lai, Steven Hsin-Yi

AU - Hsieh, S. H.

PY - 1987/12/1

Y1 - 1987/12/1

N2 - This paper presents an adaptive modeling technique for the identification and monitoring of drilling processes. The major dynamics of drill failure are captured by a threshold mixed model using the in-process drilling vibration data. The threshold learning model and the initial guess parameters are evaluated from the first drilling hole signal. The parameters are updated by an adaptive scheme in the subsequent operations. Two control indices, the transient slope and the steady-state model residuals, are estimated. The process is monitored using a forecasting control chart generated from the control parameters. The monitoring result shows that the threshold approach is capable of predicting the wear state of a tool for various machining operations with a high variability of dynamic data.

AB - This paper presents an adaptive modeling technique for the identification and monitoring of drilling processes. The major dynamics of drill failure are captured by a threshold mixed model using the in-process drilling vibration data. The threshold learning model and the initial guess parameters are evaluated from the first drilling hole signal. The parameters are updated by an adaptive scheme in the subsequent operations. Two control indices, the transient slope and the steady-state model residuals, are estimated. The process is monitored using a forecasting control chart generated from the control parameters. The monitoring result shows that the threshold approach is capable of predicting the wear state of a tool for various machining operations with a high variability of dynamic data.

UR - http://www.scopus.com/inward/record.url?scp=0023531662&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0023531662&partnerID=8YFLogxK

M3 - Paper

SP - 89

EP - 94

ER -