PSI - Issue 17

Andra Gabriela Stancu et al. / Procedia Structural Integrity 17 (2019) 238–245 A. G. Stancu et al./ Structural Integrity Procedia 00 (2019) 000 – 000

244

7

deviation of continuous monitoring dropped quickly before entering a horizontally steady line, indicating a low failure rate for the machinery.

Figure 4. Bathtub Curve: the variation of failure rate of components during their life [14].

4. Conclusion and remarks

A long-term monitoring approach was adopted and implemented by means of acoustic emission technique. The collected data was processed using a baseline and signature definition algorithm, followed by a similarity-based analysis. In summary, the baseline definition process requires, in this case, continuous data acquisition during a period of approximately 30 minutes, which will then be used as reference for the assessment of the established signatures. Minor deviation from the baseline was observed during the early stage of monitoring which increases the confidence in the use of the defined baseline. Specifically for rotating machinery, the solution presented was demonstrated successfully, using baseline acquisition and similarity analysis for long-term operation without any interruption. However, the proposed algorithm is intended generally for condition monitoring of various machineries and its use can be extended to a wide range of techniques. For any particular case, a customized approach will require identifying the relevant long-term remote condition monitoring parameters. Furthermore, other physical parameters need further consideration, which could be integrated in the monitoring process, including loading cases, operation schedule, state of lubrication, etc. Establishing a relationship between the likelihood of failure of the machinery of interest with the monitored deviation will enable operators to address efficiently their maintenance planning schemes by evaluating the remaining lifetime. Therefore, future efforts will be taken towards linking the rate of failure to the similarity analysis performed.