A systematic review of artificial neural network applications in thermal spray coatings for high-temperature turbomachinery corrosion prevention

Purpose Turbomachinery components operating at temperatures above 800°C are highly susceptible to corrosion, leading to reduced efficiency and shortened service life. Conventional thermal spray coating optimization relies heavily on trial-and-error methods, achieving limited coating uniformity (70%–80%) and restricting durability to 3–5 years. This study aims to systematically evaluate the role of artificial neural networks (ANNs) as advanced tools for optimizing thermal spray coatings to enhance corrosion resistance in high-temperature turbomachinery applications. Design/methodology/approach A comprehensive review of 45 peer-reviewed studies was conducted to analyze the application of ANN-based models, including Multilayer Perceptrons, hybrid ANN frameworks and emerging deep learning architectures such as convolutional neural networks and Long Short-Term Memory networks. The studies were critically assessed with respect to predictive accuracy, reduction in experimental trials, corrosion mitigation performance and suitability for real-time process control and digital twin integration. Findings The reviewed studies demonstrate that ANN-based optimization strategies can reduce experimental trials by approximately 40% while achieving prediction accuracies exceeding 90% (R² > 0.92). ANN-optimized coatings exhibit corrosion rate reductions of 70%–80% compared to conventional approaches. Additionally, ANN models enable real-time process monitoring, adaptive control of spray parameters and integration with digital twin frameworks for accurate life prediction. However, challenges remain regarding data set quality, model generalizability and the limited adoption of physics-informed ANN models. Originality/value This review provides a consolidated and critical assessment of ANN-driven optimization of thermal spray coatings specifically for high-temperature corrosion resistance in turbomachinery. By identifying key performance gains, emerging Industry 4.0 applications and existing research gaps, the study offers actionable insights for researchers and engineers seeking to enhance coating durability, reduce maintenance costs and accelerate the digital transformation of turbomachinery surface engineering.