Abstract:

Gearbox is the core component of various machines and vehicles, so it is necessary to monitor the performance degradation of gearbox for improving the reliability of mechanical equipment and vehicle. The use of the health condition monitoring on core components such as gearbox can reduce the economic losses due to its failure In this paper, some parameters are extracted from data of run-to-failure test, for example, kurtosis, RMS and energy. The performance degradation of the gearbox is analyzed and predicted by the monitoring the parameters. In this paper, some forecasting methods such as ARMA model, moving average line and feed forward neural network are compared by calculate absolute error.

Keywords: