ABSTRACT
High temperature exists universally, such as in main steam piping, furnaces, turbines, and jet engines. The integrity of this equipment is essential for safe and economic operation of the whole system. To keep their safety, the life prediction techniques have attracted considerable attention [1-4]. The accuracy of life prediction is influenced by many factors, including experimental data of materials, models (physical model, mathematical model, etc.) of life prediction, numerical analysis tools, service loads, environmental eects, and so forth. Errors can happen at any step. Therefore, the on-line health monitoring method becomes the only solution to guarantee the safety of critical industrial equipment [5]. The on-line health monitoring method involves observation of systems using periodically sampled dynamic response measurements from an array of sensors, the extraction of damage-sensitive features from these measurements, and the statistical analysis of these features
to determine the current state of the system health. It can detect early degradation of material properties associated with operational usage, service loads, and environmental exposure, and predict future material properties at critical locations for components, structures, and complex systems subjected to service loads and environmental exposure over time.
