Effective control of the micro- and nanostructure of thermal barrier coatings is essential to enhance the thermal radiation performance of the coating, which helps to determine the remaining service life of the coating. This paper proposed a method to measure the radiation properties of thermal barrier coatings by terahertz nondestructive testing technique, using APS-prepared thermal barrier coatings as the object of study. Radiative properties were a comprehensive set of properties characterized by the diffuse reflectance, transmittance, and absorptance of the thermal barrier coating. The coating data in actual service were obtained by scanning electron microscopy and metallographic experiments, and the data were used as the simulation model critical value. The terahertz time-domain simulation data of coatings with different microstructural features were obtained using the finite-different time-domain (FDTD) method. In simulating the real test signals, white noise with a signal-to-noise ratio of 20 dB was added, and fast Fourier transform (FFT), short-time Fourier transform (STFT), and wavelet transform (WT) were used to reduce the noise and compare their noise reduction effects. Different machine learning methods were used to build the model, including support vector machine algorithm (SVM) and k-nearest neighbor algorithm (KNN). The principal component algorithm (PCA) was used to reduce the dimensionality of terahertz time-domain data, and the SVM algorithm and KNN algorithm were optimized using the particle swarm optimization algorithm (PSO) and the ant colony optimization algorithm (ACO), respectively, to improve the robustness of the system. The K-fold cross-validation method was used to construct the model to improve the adaptability of the model. It could be clearly seen that the novel hybrid PCA-ACO-SVM model had superior prediction performance. Finally, this work proposed a novel, convenient, nondestructive, online, safe and highly accurate method for measuring the radiation performance of thermal barrier coatings, which could be used for the judgment of the service life of thermal barrier coatings.

Studying the impact of pores is crucial to enhancing the service performance of coatings, since they are a typical microstructure feature of thermal barrier coatings. In this paper, a coating prepared by the APS method was employed as the study object, and a scanning electron microscope and optical microscope were used to calculate the porosity after spraying or high-temperature exposures. Based on this, numerical calculations and simulations were used to evaluate the impacts of the pore structure and porosity on the heat conductivity and radiation characteristics of the coating. The results showed that, at high-temperature exposures, the horizontal pores inhibited thermal conductivity and radiation, but the column pores increased heat conductivity and radiation. The heat conductivity of the coating linearly decreased as the porosity increased, whereas the extinction coefficient increased, although at a slower and slower pace. When the porosity reached 15%, if the porosity was further increased, the thermal radiation energy did not change much, indicating that increasing the porosity would only block the heat radiation to a certain amount. This new and time-saving technique for materials research utilizing simulation and numerical computing may be utilized to optimize the microstructure of coatings to increase their service performance.

The spraying process affects the microstructure and service life of thermal barrier coatings. In this work, by changing the plasma-spraying process, the second-phase micron-agglomerated particles were introduced to prepare embedded micron-agglomerated particle thermal barrier coatings. Conventional thermal barrier coatings and embedded micron-agglomerated particle thermal barrier coatings were prepared by setting spraying process parameters with different powder feeding rates and distances between powder feeders. Sintering experiments and thermal cycling experiments were carried out on conventional thermal barrier coatings and embedded micron-agglomerated particle thermal barrier coatings. The effects of spraying process parameters on the microstructure and thermal cycle life of embedded micron-agglomerated particle thermal barrier coatings were systematically studied. Three machine learning models of BP, SVM, and GA-SVM were established to evaluate the relationship between spraying parameters and coating microstructure and thermal cycle life. The results show that the sintering resistance of the new thermal barrier coating is 500% higher than that of the conventional thermal barrier coating. When the distance between the two powder feeders is 35 mm and the powder feeder 1 rate is 12 g/min, the thermal cycle life of the coating is the best. When using machine learning model evaluation, the evaluation results of the three machine learning models have certain accuracy. Among them, the GA-SVM machine learning model has the best prediction effect and the smallest prediction error. The results of this paper provide a new paradigm for further preparation of thermal barrier coatings with high performance and long life.

The distribution morphology and density of micro-agglomerated particles are the main microstructural characteristics of embedded micro-agglomerated particle thermal barrier coatings. The study of their effect on the sintering resistance of coatings can help to further improve the service life of thermal barrier coatings. Strain tolerance and thermal insulation performance are important evaluation indicators for the sintering resistance of thermal barrier coatings. In this study, embedded micro-agglomerated particle thermal barrier coatings were prepared by plasma spraying, and the distribution morphology and density of micro-agglomerated particles were analyzed and counted. Different simulation models were established to analyze the compressive stress and thermal stress of the coating, as well as the influence of the microstructure characteristics on the strain tolerance and thermal insulation performance of the coating. A machine learning model was established to evaluate the nonlinear relationship between the microstructure characteristics of the coating and its strain tolerance and thermal insulation performance. The results show that the horizontal angle of the micro-agglomerated particles in the coating has the most significant effect on the sintering resistance and is predicted using the PSO-SVM machine learning model. The predicted effects are the most important, and the coefficients of determination for strain tolerance and thermal insulation are as high as 0.988 and 0.945, respectively, indicating a strong correlation between the predicted and actual values. This research technique used experimental research-simulation computation-machine learning can be used to optimize the microstructure of coatings and guide the preparation of high-performance thermal barrier coatings.