A methodology using health and usage monitoring system data for payload life prediction
This paper presents a methodology to monitor the fatigue life of aerospace structures and hence the remaining allowable fatigue life. In fatigue clearance, conservative load assumptions are made. However, in reality, a structure may see much lower loads and so would be usable for much longer. An example ofthis is air carried guided missiles. In the UK, missiles must be decommissioned after a period of carriage. The implementation of a system that can monitor the usage of a missile during its time in service is advantageous to the military customer and provides a competitive advantage for the missile manufacture inexport markets where reduced through-life costs, longer in-service lives and increased safety are desired. The proposed methodology provides a means to monitor the service life of a missile. This paper describes how machine learning algorithms can be used with accelerometers to determine loads on a missile structure which would then be used to predict how long the missile has left in service.