The core task of intelligent detection in electrified railway catenaries is to detect their supporting components. However, due to the large number and variety of components in catenary inspection images, labeled catenary data is often limited, and few studies have focused on leveraging large amounts of unlabeled datasets in this field. This article proposes a novel self-supervised pretraining model catenary support rod masking-based masked image modeling (CSRM-MIM) for detecting catenary support components (CSCs), which effectively utilizes the valuable information in unlabeled catenary data. Specifically, a new semantic catenary support rod area masking (CSRA masking) strategy based on the CSRA features is proposed to guide the model in learning meaningful catenary semantic information during the pretraining process. Additionally, a new Siamese pretrained network framework is designed, incorporating an information interaction enhancement module (IIEM) and a dual reconstruction network to perform dual reconstruction on the semantic mask, thereby extracting global features in the catenary domain. Finally, a multiscale knowledge distillation (MSKD) strategy optimized by the cross-layer fusion module cross-layer fusion decoding (CLFD) is introduced to assist the self-supervised model in acquiring specific-general representations for the catenary field and transferring multiscale information, which benefits subsequent detection tasks. Experimental results demonstrate that this method significantly enhances the detector’s performance in identifying CSCs, owing to the retained self-supervised learning (SSL) strategy.

Overhead conductor rail (OCR) is an important power supply device in tunnel railway lines. The penalty method is widely adopted to model the dynamic interaction behaviour of the pantograph-OCR system (POCRs), while it is difficult to determine a proper contact stiffness value. This paper proposes a three-dimensional contact formulation for the POCRs based on the Lagrange multiplier method. Accurate contact forces can be obtained easily by this contact formulation without the trouble of determining the contact stiffness. Contact force is defined in the global coordinate system as a three-dimensional vector. Coulomb law of friction is adopted in this paper to describe the tangential contact force, and the train speed is included in the contact formulation. The central finite difference approximation method is utilised, and an iteration procedure is elaborated to calculate the accurate contact force vector. The proposed contact formulation is verified by comparing simulation results with in-line measured data. The maximum statistical relative error of the contact force is only 2.63%. The advantages of the proposed contact formulation are also demonstrated by comparing it with the penalty method.

Overhead Conductor Rail (OCR) is a power supply device in tunnel railways. This paper aims to investigate the influence of train speed and support stiffness on the pantograph-OCR system (POCRs) dynamic behaviour, and propose improvement measures to the existing power feeding system to accommodate higher speeds. Firstly, a field test is conducted to measure the equivalent stiffness of the supports currently utilised in the OCR system. The test results are adopted to modify the OCR model. Subsequently, a lumped mass model of the pantograph and the OCR model considering interlocking joints are developed. The analysis reveals that the contact force exhibits a nonlinear variation as the support stiffness increases, and good dynamic interaction can be achieved when support stiffness falls between 10 ⁴ and 10 ⁵ N/m. Then time-frequency analysis explains the origin and amplification factors of multiple frequency components of the contact forces. The results show that structural wavelengths such as span length, half span length and interlocking joint interval are the primary sources of contact force components. Finally, the resonance mechanism of the POCRs is revealed, explaining the occurrence of deteriorated interaction performance under specific support stiffness.

Overhead conductor rail (OCR) is widely used in tunnel railway lines to provide continuous energy for electric trains. Current studies mainly focus on the interaction of the low-speed pantograph–OCR system (POCRs), whereas the increase in speed poses new challenges in the power supply of the locomotives. This article develops a mathematical model based on the absolute nodal coordinate formulation for simulating the dynamic behavior of POCRs that operates at high speed, integrated with the contact wire irregularities (CWIs). The presented model is verified by comparing it with the measured data, and the maximum statistical relative error of the contact force is only 7.52%. Next, the sensitivity analysis is performed to reveal the effect of critical parameters of the OCR on the contact force at 200 km/h. Furthermore, the resonance mechanism of the POCRs is revealed based on a simplified model, which can be used to explain the appearance of bad current collection quality with some specific span lengths and CWI wavelengths. Finally, a linearized simulation algorithm is presented, saving about 99.65% of the computation time. The comparative analysis indicates that the geometrical nonlinearity of OCR can be neglected for improving computational efficiency.

The rigid catenary system is widely used in tunnels to power electric trains via contact with a pantograph. Due to gravity, the contact wire normally has a sag that may affect the dynamic interaction performance with a pantograph. To reduce the contact wire sag, the most efficient measure is to improve the moment of inertia of the conductor rail, which is used to clamp the contact wire. Six new types of conductor rail with large moments of inertia are developed based on a conventional conductor rail. Then both the static and dynamic analyses are conducted to investigate the performance of the new types of conductor rail with a big moment of inertia. The conductor rail’s 3D solid finite element model is built using a finite element approach. The vertical deflection and the stress distribution are comparatively analyzed among different types of conductor rail. The analysis results indicate that the vertical deflection and maximum stress are significantly reduced when using the conductor rail with a large moment of inertia. The best performance is observed when the conductor rail of case 1 is used. The maximum sag is reduced by 28.37%, and the maximum stress is decreased by 27.76% compared with the conventional conductor. Finally, a pantograph model is included to evaluate the dynamic performance of the conductor rail with large moments of inertia. The results indicate that contact force fluctuation is significantly reduced after the conductor rails with large moments of inertia are presented. The conductor rail of case 1 shows the best performance, which can reduce the contact force standard deviation by 32% and 27% at speeds of 160 km/h and 200 km/h.

This study focuses on developing a comprehensive model of a rigid overhead system, which includes essential components such as the suspension structure, positioning clamp, and expansion joint. The modelling approach utilizes finite element theory and beam elements to accurately represent the displacement, stiffness, and mass characteristics of the system. The models also incorporate the suspension structure and positioning line clamp, which play crucial roles in suspending and positioning the busbar. Various suspension structures and positioning line clamps are evaluated based on their dynamic characteristics. The expansion joint, responsible for connecting different anchor sections of the rigid overhead system, undergoes a detailed analysis. Different assembly scenarios, including ideal and deflected assembly conditions, are considered. To simulate the dynamic behaviour of the expansion joint, additional beams are introduced into the system model. The primary finding of the analysis is that the maximum stresses observed in the constructed expansion joint model, under different temperature conditions and normal/deflected assembly conditions, remain within the permissible stress limits of the material. This indicates a high level of safety. However, certain areas exhibit stress concentration, particularly at the sliding block B and sliding rod A positions. This stress concentration is primarily attributed to the unique assembly form of the expansion joint. To improve stress distribution and enhance service reliability, the analysis suggests optimizing the installation deflection angle and geometric design of the expansion joint. Furthermore, the concentrated mass at the expansion joint significantly impacts the current collection quality of the pantograph-overhead system. Mitigating this negative impact can be achieved by reducing the mass of the expansion joint.