This paper presents a non-contact measurement of the realistic catenary geometry deviation in the Norwegian railway network through a laser rangefinder. The random geometry deviation is included in the catenary model to investigate its effect on the pantograph–catenary interaction. The dispersion of the longitudinal deviation is assumed to follow a Gaussian distribution. A power spectrum density represents the vertical deviation in the contact wire. Based on the Monte Carlo method, several geometry deviation samples are generated and included in the catenary model. A lumped mass pantograph with flexible collectors is employed to reproduce the high-frequency behaviours. The stochastic analysis results indicate that the catenary geometry deviation causes a significant dispersion of the pantograph–catenary interaction response. The contact force standard deviations measured by the inspection vehicle are within the scope of the simulation results. A critical cut-off frequency that covers 1/16 of the dropper interval is suggested to fully describe the effect of the catenary geometry deviation on the contact force. The statistical minimum contact force is recommended to be modified according to the tolerant contact loss rate at high frequency. An unpleasant interaction performance of the pantograph–catenary can be expected at the catenary top speed when the random catenary geometry deviation is included.

Corrosion is one of the main damage mechanisms in civil engineering structures today. Rapid identification and accurate assessment of corrosion in structures are essential to ensure the efficient allocation of limited funds for the maintenance and renewal of existing structures. Vision-based neural networks have been widely used in corrosion detection, in which convolutional neural network (CNN)-like models remain dominant. However, these conventional network models exhibit a saturating performance. Because of the self-attention mechanism, the transformer is the newest breakthrough in computer vision and is becoming state of the art. As the complexity of structures increases, transformer-based methods have no saturating performance. This study proposes a corrosion localization and evaluation architecture for a larger view based on semantic segmentation and image stitching for automatic localization and diagnosis of corrosion from stitched images. The experimental results showed that the proposed method achieved better corrosion detection performance (F1-score = 68.2%) than that of the mainstream CNN-like models U-Net (F1-score = 61.8%) and DeepLabV3 + (F1-score = 60.1%). Image stitching is utilized for corrosion segmentation in larger view images, and the field test shows that the proposed architecture could stitch corrosion prediction from different images.