TY - GEN
T1 - Complex Industrial Machinery Health Diagnosis Challenges and Strategies
AU - Wang, Hsiao Yu
AU - Hung, Ching Hua
N1 - Publisher Copyright:
© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2024.
PY - 2024
Y1 - 2024
N2 - This study is dedicated to addressing a spectrum of pivotal challenges and predicting their potential ramifications. Specifically, its objectives encompass the detection of tool breakage in milling-turning composite machinery, the assessment of the service life of punching machine heads, and the evaluation of mold longevity in forging apparatus, among other intricacies. The overarching objective is the establishment of an equipment health diagnosis system tailored for intricate industrial setups. It is evident from our interactions with the industry that the rationale for monitoring strategies and threshold values are contingent upon the idiosyncratic attributes of the equipment and the sector. While the metal processing sector has been trailing behind the semiconductor industry in the realm of intelligent monitoring by an approximate span of a decade, it faces an analogous array of challenges. These encompass dwindling demographics, leading to an increased reliance on external labor for shifts, elevated personnel turnover rates thereby limiting the availability of experienced personnel for tasks such as tool changes, mold replacements, and maintenance. Additionally, the necessity to uphold traceability standards for mold and punching head usage history, notably in the context of aerospace industry compliance, compounds these challenges. Consequently, the industry aspires to achieve two paramount objectives for vital production equipment: first, the execution of failure diagnostics to appraise tool or mold longevity and assess product quality. Second, the transition from time-based to condition-based maintenance practices, even under conditions that necessitate frequent mold substitutions to cater to diverse product manufacturing needs.
AB - This study is dedicated to addressing a spectrum of pivotal challenges and predicting their potential ramifications. Specifically, its objectives encompass the detection of tool breakage in milling-turning composite machinery, the assessment of the service life of punching machine heads, and the evaluation of mold longevity in forging apparatus, among other intricacies. The overarching objective is the establishment of an equipment health diagnosis system tailored for intricate industrial setups. It is evident from our interactions with the industry that the rationale for monitoring strategies and threshold values are contingent upon the idiosyncratic attributes of the equipment and the sector. While the metal processing sector has been trailing behind the semiconductor industry in the realm of intelligent monitoring by an approximate span of a decade, it faces an analogous array of challenges. These encompass dwindling demographics, leading to an increased reliance on external labor for shifts, elevated personnel turnover rates thereby limiting the availability of experienced personnel for tasks such as tool changes, mold replacements, and maintenance. Additionally, the necessity to uphold traceability standards for mold and punching head usage history, notably in the context of aerospace industry compliance, compounds these challenges. Consequently, the industry aspires to achieve two paramount objectives for vital production equipment: first, the execution of failure diagnostics to appraise tool or mold longevity and assess product quality. Second, the transition from time-based to condition-based maintenance practices, even under conditions that necessitate frequent mold substitutions to cater to diverse product manufacturing needs.
KW - Bayesian regularization
KW - Ensemble Empirical Mode Decomposition
KW - Root Mean Square
KW - Statistical overlap factor
UR - http://www.scopus.com/inward/record.url?scp=85189630084&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-55976-1_13
DO - 10.1007/978-3-031-55976-1_13
M3 - Conference contribution
AN - SCOPUS:85189630084
SN - 9783031559754
T3 - Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST
SP - 130
EP - 140
BT - Smart Grid and Internet of Things - 7th EAI International Conference, SGIoT 2023, Proceedings
A2 - Deng, Der-Jiunn
A2 - Chen, Jyh-Cheng
PB - Springer Science and Business Media Deutschland GmbH
T2 - 7th EAI International Conference on Smart Grid and Internet of Things, SGIoT 2023
Y2 - 18 November 2023 through 19 November 2023
ER -