Typical Damage Prediction and Reliability Analysis of Superheater Tubes in Power Station Boilers Based on Multisource Data Analysis

The superheater and re-heater piping components in supercritical thermal power units are prone to creep fatigue failure fracture after extensive use due the high pressure temperature environment. Therefore, safety assessment for superheaters re-heaters such an environment is critical. However, actual service operation data frequently insufficient, resulting low accuracy of assessment. Based on problems, order address issues susceptibility creep, inaccurate fracture, creep–fatigue coupling rupture a assessment, their remaining life prediction reliability, as well lack data, multisource heterogeneous generated from plants combined with deep learning technology was used this paper. As such, three real-time operating conditions’ (temperature, pressure, stress amplitude) during equipment predicted by training architecture long short-term memory (LSTM) neural network suitable processing time-series backpropagation through time (BPTT) algorithm optimize model compared physical model. Damage final tubes station boilers carried out. Weibull distribution obtain trend cumulative risk change assess predict condition overall system pressurized boilers.