For efficient maintenance of critical industrial asset one requires an estimate of the current condition as well as an estimate of the remaining useful life. One can hence guarantee timely fault accommodation and avoid unanticipated breakdowns and equipment failures. The paper presents a concept of a system for on-line diagnostics and prognostics of asset condition whose aim is to make implementation easier, cost effective and highly portable. The system builds on a distributed sensor network performing signal acquisition, local signal processing and fusion of the diagnostic results on a central server. Hence the state of health as well as the estimate of the remaining useful life can be generated online. Key to the concept is optimal data storage and data manipulation framework with built-in condition monitoring taxonomy. To address all these issues the potential of the MIMOSA OSA-EAI standard are exploited. Due to its thoroughness MIMOSA OSA-EAI is unjustly deemed cumbersome and is either avoided or neglected. To bring it closer to the practitioners its main features are carefully addressed. The implementation of a platform prototype on a large milling machine is described.
[1]
Joseph Mathew,et al.
Rotating machinery prognostics. State of the art, challenges and opportunities
,
2009
.
[2]
Pavle Boškoski,et al.
Fault detection of mechanical drives under variable operating conditions based on wavelet packet Rényi entropy signatures
,
2012
.
[3]
Benoît Iung,et al.
Maintenance alternative integration to prognosis process engineering
,
2007
.
[4]
A. Rakar,et al.
Diagnostic reasoning under conflicting data: the application of the transferable belief model
,
2002
.
[5]
M. Gasperin,et al.
Prediction of the remaining useful life: An integrated framework for model estimation and failure prognostics
,
2012,
2012 IEEE Conference on Prognostics and Health Management.
[6]
Benoît Iung,et al.
Conceptual framework for e-Maintenance: Illustration by e-Maintenance technologies and platforms
,
2009,
Annu. Rev. Control..