International Conference on Mechanical Engineering and Technology (ICMET-London 2011)
74 Novelty Detection for Predictive Maintenance Scheduling for Industrial Gas Turbines
Download citation file:
- Ris (Zotero)
- Reference Manager
The paper presents results of an investigation to predict impending failure mechanisms of a gearbox drive train in the sub 15MW class of the Siemens gas turbine product range. Particular emphasis is given to the prediction of gearbox failures and inter-connected components. Experimental results from real-time data show that the application of SVM techniques provides an efficient basis for minimising the impact of unscheduled maintenance requirements, on product lifetime and cost for these units.