Detection and diagnosis of material degradation are of a complex and challenging task since it is presently hand-operated by a human. Therefore, it leads to misinterpretation and avoids correct classification and diagnosis. In this paper, we develop a computer-assisted detection method of material failure by utilizing a deep learning approach. A deep convolutional neural network (CNN) model, combined with an image processing technique, e.g., adaptive histogram equalization, is trained to classify a real-world turbine tube degradation image data set, which is retrieved from a power generation company. The experimental result demonstrates the effectiveness of the proposed approach with predictive classification accuracy is up to 99.99% in comparison with a shallow machine learning algorithm, e.g., linear SVM. Furthermore, performance evaluation of a deep CNN with and without an above-mentioned image processing technique is exhibited and benchmarked. We successfully demonstrate a novel application in constructing a deep-structure neural network model for material degradation diagnosis, which is not available in the current literature.
We developed a health-monitoring methodology for high-temperature steam pipes that estimated the life prediction of creep–fatigue interaction by directly measuring the displacement of hot parts. Three different methods (boiler code, design stress, and operating stress) were used to estimate the stress of the high-temperature pipe system. As a theoretical approach, the German boiler standard code calculates the stress according to the pipe shape, while design stress, which is also called allowable stress, was determined by a function of the operating temperature. The operating stress was immediately calculated using the surrogate model, with maximum displacement measured using the 3D displacement measurement system. To achieve the surrogate model, the stress was estimated by the pipe-stress analysis under the given displacements, and the surface-response model was developed to relate the stress and displacement. We showed that those methods are efficient methods to predict the stress and are applicable in health-monitoring methodology. Finally, the creep life and the low-cycle fatigue life were investigated using the Larson–Miller parameter equation, as well as the Smith, Hirschberg, and Manson equations. Our proposed monitoring system can be used to predict the fatigue and creep life of high-temperature steam pipes in real time, and we believe that the system can be applied to actual maintenance in thermal power plants.