A Real-Time Condition Monitoring System by using Seasonal ARIMA Model and Control Charting
Abstract
En
This paper is concerned with research on early-failure monitoring systems for safety of railway systems. The work presented here have led to the development of ideas and techniques for the employment of time series modelling and control charting for on-line temperature monitoring of railcar brakes. A software package implementing the real-time monitoring scheme is presented. The temperature signal is sampled and the readings are filtered using a time-series model. In particular, a seasonal ARIMA model is exploited. The filtered signal, which has well defined statistical properties, is then plotted against proper control limits. The motivation of the research is the need for improved reliability of equipment and quality of service to metro passengers.
This paper is concerned with research on early-failure monitoring systems for safety of railway systems. The work presented here have led to the development of ideas and techniques for the employment of time series modelling and control charting for on-line temperature monitoring of railcar brakes. A software package implementing the real-time monitoring scheme is presented. The temperature signal is sampled and the readings are filtered using a time-series model. In particular, a seasonal ARIMA model is exploited. The filtered signal, which has well defined statistical properties, is then plotted against proper control limits. The motivation of the research is the need for improved reliability of equipment and quality of service to metro passengers.
DOI Code:
�
Keywords:
Seasonal ARIMA model; SPC; FVC; SCC
Full Text: PDF
